@opengeni/config 1.0.4 → 1.1.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/index.js CHANGED
@@ -1,13 +1,24 @@
1
+ // src/managed-user-admission.ts
2
+ function managedUserEmailAllowed(allowedUserEmails, email) {
3
+ return allowedUserEmails === void 0 || typeof email === "string" && allowedUserEmails.some((allowed2) => allowed2.toLowerCase() === email.trim().toLowerCase());
4
+ }
5
+
1
6
  // src/index.ts
7
+ import { isRetiredNativeAtlassianTool } from "@opengeni/contracts/atlassian-native-retirement";
2
8
  import {
9
+ directModelConnectionSpec,
3
10
  BillingMode,
4
11
  CAPABILITY_DESCRIPTORS,
12
+ agentConfigDeploymentLimitsFromAllowlist,
5
13
  DEFAULT_FIRST_PARTY_MCP_TOOLS,
14
+ DEFAULT_OPENGENI_DOCUMENTATION_URL,
15
+ currentAgentLearningToolSelection,
6
16
  Entitlements,
7
17
  EntitlementsMode,
8
18
  KnowledgeSourceSyncLimits,
9
19
  LatencyMode,
10
20
  MAX_NESTED_AGENT_DEPTH,
21
+ McpServerConnectionRef as PublicMcpServerConnectionRef,
11
22
  ProductAccessMode,
12
23
  ReasoningEffort,
13
24
  FIRST_PARTY_MCP_TOOL_NAMES,
@@ -42,7 +53,365 @@ import {
42
53
  } from "@opengeni/xai-subscription";
43
54
  import { XAI_SUBSCRIPTION_MODEL_ID_PREFIX as XAI_SUBSCRIPTION_MODEL_ID_PREFIX2 } from "@opengeni/xai-subscription";
44
55
  import { createHash } from "crypto";
56
+ import { isIP } from "net";
57
+ import { z as z2 } from "zod";
58
+
59
+ // src/claude-subscription-usage.ts
60
+ import {
61
+ ClaudeUsageWindowId
62
+ } from "@opengeni/contracts";
63
+ var prefixes = {
64
+ five_hour: ["5h"],
65
+ seven_day: ["7d"],
66
+ seven_day_opus: ["7d-opus"],
67
+ seven_day_sonnet: ["7d-sonnet"],
68
+ seven_day_overage_included: ["7d_oi", "7d-overage-included"],
69
+ overage: ["overage"]
70
+ };
71
+ function percent(value, scale) {
72
+ if (typeof value !== "number" && (typeof value !== "string" || !value.trim())) return null;
73
+ const number = Number(value) * scale;
74
+ return Number.isFinite(number) && number >= 0 ? number : null;
75
+ }
76
+ function epoch(value) {
77
+ if (!value || !/^\d+(?:\.\d+)?$/.test(value)) return null;
78
+ const time = Number(value) * 1e3;
79
+ return Number.isFinite(time) && time > 0 && time < 864e13 ? new Date(time).toISOString() : null;
80
+ }
81
+ function status(value) {
82
+ return value === "allowed" || value === "allowed_warning" || value === "rejected" ? value : null;
83
+ }
84
+ function latestReset(left, right, observedAt) {
85
+ return new Date(
86
+ Math.max(
87
+ left ? Date.parse(left) : Date.parse(observedAt) + 6e4,
88
+ right ? Date.parse(right) : Date.parse(observedAt) + 6e4
89
+ )
90
+ ).toISOString();
91
+ }
92
+ function evidenceKey(item) {
93
+ return JSON.stringify([
94
+ item.upstreamModelId,
95
+ item.upstreamModelId ? null : item.representativeClaim,
96
+ item.source === "provider"
97
+ ]);
98
+ }
99
+ function tiedRequest(left, right) {
100
+ const denied = left.status === "rejected" ? left : right.status === "rejected" ? right : null;
101
+ return {
102
+ ...denied ?? left,
103
+ resetsAt: denied ? latestReset(
104
+ left.status === "rejected" ? left.resetsAt : denied.resetsAt,
105
+ right.status === "rejected" ? right.resetsAt : denied.resetsAt,
106
+ left.observedAt
107
+ ) : null,
108
+ representativeClaim: left.representativeClaim === right.representativeClaim ? left.representativeClaim : null,
109
+ status: denied ? "rejected" : null,
110
+ overageStatus: null,
111
+ overageResetsAt: null
112
+ };
113
+ }
114
+ function parseClaudeUsageHeaders(headers, now = /* @__PURE__ */ new Date(), upstreamModelId) {
115
+ const observedAt = now.toISOString();
116
+ const claim = ClaudeUsageWindowId.safeParse(
117
+ headers.get("anthropic-ratelimit-unified-representative-claim")
118
+ );
119
+ const claimedStatus = status(headers.get("anthropic-ratelimit-unified-status"));
120
+ const claimedReset = epoch(headers.get("anthropic-ratelimit-unified-reset"));
121
+ const overageStatus = status(headers.get("anthropic-ratelimit-unified-overage-status"));
122
+ const overageResetsAt = epoch(headers.get("anthropic-ratelimit-unified-overage-reset"));
123
+ const windows = ClaudeUsageWindowId.options.flatMap((id) => {
124
+ const header = (suffix) => prefixes[id].map(
125
+ (prefix) => headers.get(`anthropic-ratelimit-unified-${prefix}-${suffix}`)
126
+ ).find((value) => value !== null) ?? null;
127
+ const representative = claim.success && claim.data === id;
128
+ const usedPercent = percent(header("utilization"), 100);
129
+ const resetsAt = epoch(header("reset")) ?? (representative ? claimedReset : null);
130
+ const windowStatus = status(header("status")) ?? (representative ? claimedStatus : null);
131
+ if (usedPercent === null && resetsAt === null && windowStatus === null) return [];
132
+ return [{ id, usedPercent, resetsAt, status: windowStatus, observedAt }];
133
+ });
134
+ return windows.length || claimedStatus || overageStatus ? {
135
+ windows,
136
+ observedAt,
137
+ source: "response_headers",
138
+ requestStatus: {
139
+ status: claimedStatus,
140
+ resetsAt: claimedReset,
141
+ representativeClaim: claim.success ? claim.data : null,
142
+ overageStatus,
143
+ overageResetsAt,
144
+ upstreamModelId: upstreamModelId ?? null,
145
+ observedAt
146
+ }
147
+ } : null;
148
+ }
149
+ function parseClaudeUsageResponse(value, now = /* @__PURE__ */ new Date()) {
150
+ if (!value || typeof value !== "object" || Array.isArray(value)) return null;
151
+ const observedAt = now.toISOString();
152
+ const windows = ClaudeUsageWindowId.options.flatMap((id) => {
153
+ const window = value[id];
154
+ if (!window || typeof window !== "object" || Array.isArray(window)) return [];
155
+ const fields = window;
156
+ const usedPercent = percent(fields.utilization, 1);
157
+ const time = typeof fields.resets_at === "string" ? Date.parse(fields.resets_at) : NaN;
158
+ const resetsAt = Number.isFinite(time) ? new Date(time).toISOString() : null;
159
+ if (usedPercent === null && resetsAt === null) return [];
160
+ return [{ id, usedPercent, resetsAt, status: null, observedAt }];
161
+ });
162
+ return windows.length ? { windows, observedAt, source: "provider" } : null;
163
+ }
164
+ function emptyClaudeUsage(credentialVersion) {
165
+ return {
166
+ connected: credentialVersion !== null,
167
+ credentialVersion,
168
+ windows: [],
169
+ observedAt: null,
170
+ source: null,
171
+ refreshStatus: "not_checked",
172
+ refreshCheckedAt: null
173
+ };
174
+ }
175
+ function mergeClaudeUsage(current, observation) {
176
+ const windows = new Map(current.windows.map((window) => [window.id, window]));
177
+ for (const incomingWindow of observation.windows) {
178
+ const window = { ...incomingWindow, source: incomingWindow.source ?? observation.source };
179
+ const previous = windows.get(window.id);
180
+ if (!previous || previous.observedAt < window.observedAt) windows.set(window.id, window);
181
+ else if (previous.observedAt === window.observedAt && JSON.stringify(previous) !== JSON.stringify(window)) {
182
+ const rejected = previous.status === "rejected" || window.status === "rejected";
183
+ const full = (previous.usedPercent ?? 0) >= 100 || (window.usedPercent ?? 0) >= 100;
184
+ windows.set(window.id, {
185
+ ...previous,
186
+ usedPercent: previous.usedPercent === null ? window.usedPercent : window.usedPercent === null ? previous.usedPercent : Math.max(previous.usedPercent, window.usedPercent),
187
+ status: rejected ? "rejected" : previous.status === window.status ? previous.status : null,
188
+ resetsAt: rejected || full ? latestReset(previous.resetsAt, window.resetsAt, window.observedAt) : previous.resetsAt === window.resetsAt ? previous.resetsAt : null,
189
+ source: previous.source === "response_headers" || window.source === "response_headers" ? "response_headers" : "provider"
190
+ });
191
+ }
192
+ }
193
+ const incoming = observation.requestStatus;
194
+ const restrictions = new Map(
195
+ (current.requestRestrictions ?? []).map((item) => [evidenceKey(item), item])
196
+ );
197
+ for (const item of [
198
+ current.requestStatus?.status ? current.requestStatus : null,
199
+ ...observation.requestRestrictions ?? [],
200
+ incoming?.status ? incoming : null
201
+ ]) {
202
+ if (!item) continue;
203
+ const key = evidenceKey(item);
204
+ const previous = restrictions.get(key);
205
+ const sanitized = {
206
+ ...item,
207
+ overageStatus: null,
208
+ overageResetsAt: null
209
+ };
210
+ if (!previous || item.observedAt > previous.observedAt) restrictions.set(key, sanitized);
211
+ else if (item.observedAt === previous.observedAt && JSON.stringify(previous) !== JSON.stringify(sanitized))
212
+ restrictions.set(key, tiedRequest(previous, sanitized));
213
+ }
214
+ const directWindows = [
215
+ ...current.windows.filter(
216
+ (window) => window.source === "provider" || !window.source && current.source === "provider" && current.observedAt === window.observedAt
217
+ ),
218
+ ...observation.windows.filter((window) => (window.source ?? observation.source) === "provider")
219
+ ];
220
+ for (const window of directWindows) {
221
+ if (window.usedPercent === null || window.usedPercent >= 100) continue;
222
+ const proof = {
223
+ status: null,
224
+ resetsAt: null,
225
+ representativeClaim: window.id,
226
+ upstreamModelId: null,
227
+ overageStatus: null,
228
+ overageResetsAt: null,
229
+ observedAt: window.observedAt,
230
+ source: "provider"
231
+ };
232
+ const key = evidenceKey(proof);
233
+ if (!restrictions.has(key) || restrictions.get(key).observedAt < proof.observedAt)
234
+ restrictions.set(key, proof);
235
+ }
236
+ for (const [key, item] of restrictions) {
237
+ if (item.status !== "rejected") continue;
238
+ const proof = [...restrictions.values()].find(
239
+ (candidate) => candidate.source === "provider" && candidate.representativeClaim === item.representativeClaim && candidate.observedAt > item.observedAt && candidate.representativeClaim !== "overage" && candidate.representativeClaim !== "seven_day_overage_included"
240
+ );
241
+ if (proof)
242
+ restrictions.set(key, {
243
+ ...item,
244
+ status: null,
245
+ resetsAt: null,
246
+ observedAt: proof.observedAt
247
+ });
248
+ }
249
+ let requestRestrictions = [...restrictions.values()];
250
+ const evidenceTime = requestRestrictions.reduce(
251
+ (value, item) => item.observedAt > value ? item.observedAt : value,
252
+ observation.observedAt
253
+ );
254
+ const liveDenials = requestRestrictions.filter(
255
+ (item) => item.status === "rejected" && (item.resetsAt ? Date.parse(item.resetsAt) : Date.parse(item.observedAt) + 6e4) > Date.parse(evidenceTime)
256
+ );
257
+ if (requestRestrictions.length > 64 && liveDenials.length === 0) {
258
+ requestRestrictions = requestRestrictions.sort((left, right) => right.observedAt.localeCompare(left.observedAt)).slice(0, 64);
259
+ } else if (requestRestrictions.length > 64) {
260
+ const observedAt = evidenceTime;
261
+ const resetsAt = new Date(
262
+ Math.max(
263
+ ...liveDenials.map(
264
+ (item) => item.resetsAt ? Date.parse(item.resetsAt) : Date.parse(item.observedAt) + 6e4
265
+ ),
266
+ Date.parse(observedAt) + 6e4
267
+ )
268
+ ).toISOString();
269
+ requestRestrictions = [
270
+ {
271
+ status: "rejected",
272
+ resetsAt,
273
+ representativeClaim: null,
274
+ overageStatus: null,
275
+ overageResetsAt: null,
276
+ upstreamModelId: null,
277
+ observedAt
278
+ }
279
+ ];
280
+ }
281
+ const newest = !current.observedAt || current.observedAt < observation.observedAt;
282
+ const tied = current.observedAt === observation.observedAt;
283
+ let requestStatus = current.requestStatus ?? null;
284
+ if (newest) requestStatus = incoming ?? null;
285
+ else if (tied && JSON.stringify(requestStatus) !== JSON.stringify(incoming ?? null)) {
286
+ requestStatus = incoming && requestStatus && evidenceKey(incoming) === evidenceKey(requestStatus) ? tiedRequest(requestStatus, incoming) : null;
287
+ }
288
+ return {
289
+ ...current,
290
+ requestRestrictions,
291
+ requestStatus,
292
+ windows: ClaudeUsageWindowId.options.flatMap(
293
+ (id) => windows.has(id) ? [windows.get(id)] : []
294
+ ),
295
+ ...newest ? {
296
+ observedAt: observation.observedAt,
297
+ source: observation.source
298
+ // A newer direct or partial observation cannot retain an older paid
299
+ // fallback authorization. Window readings remain independently merged.
300
+ } : {}
301
+ };
302
+ }
303
+
304
+ // src/claude-subscription-oauth.ts
45
305
  import { z } from "zod";
306
+ var CLAUDE_OAUTH_CLIENT_ID = "9d1c250a-e61b-44d9-88ed-5944d1962f5e";
307
+ var CLAUDE_OAUTH_AUTHORIZE_URL = "https://claude.com/cai/oauth/authorize";
308
+ var CLAUDE_OAUTH_TOKEN_URL = "https://platform.claude.com/v1/oauth/token";
309
+ var CLAUDE_OAUTH_REDIRECT_URL = "https://platform.claude.com/oauth/code/callback";
310
+ var CLAUDE_OAUTH_SCOPES = ["user:inference", "user:profile"];
311
+ var ClaudeOAuthTokenResponse = z.object({
312
+ access_token: z.string().max(16384).regex(/^sk-ant-oat[0-9]+-\S+$/),
313
+ refresh_token: z.string().min(1).max(16384).optional(),
314
+ expires_in: z.number().int().positive().max(31536e3),
315
+ scope: z.string().min(1).max(8192)
316
+ }).passthrough();
317
+ function claudeAuthorizationUrl(input) {
318
+ const url = new URL(CLAUDE_OAUTH_AUTHORIZE_URL);
319
+ url.searchParams.set("code", "true");
320
+ url.searchParams.set("client_id", CLAUDE_OAUTH_CLIENT_ID);
321
+ url.searchParams.set("response_type", "code");
322
+ url.searchParams.set("redirect_uri", CLAUDE_OAUTH_REDIRECT_URL);
323
+ url.searchParams.set("scope", CLAUDE_OAUTH_SCOPES.join(" "));
324
+ url.searchParams.set("code_challenge", input.challenge);
325
+ url.searchParams.set("code_challenge_method", "S256");
326
+ url.searchParams.set("state", input.state);
327
+ return url.toString();
328
+ }
329
+
330
+ // src/subscription-account-selection.ts
331
+ function subscriptionAccountShardIndex(sessionId, candidateCount) {
332
+ if (!Number.isSafeInteger(candidateCount) || candidateCount <= 0)
333
+ throw new Error("Subscription shard candidate count must be positive");
334
+ let hash = 2166136261;
335
+ for (let index = 0; index < sessionId.length; index += 1) {
336
+ hash ^= sessionId.charCodeAt(index);
337
+ hash = Math.imul(hash, 16777619);
338
+ }
339
+ return (hash >>> 0) % candidateCount;
340
+ }
341
+ function selectSubscriptionAccount(input) {
342
+ const pinned = input.eligible.find((account) => account.id === input.pinnedCredentialId);
343
+ if (input.pinnedCredentialId && input.pinSource !== "policy") return pinned ?? null;
344
+ if (!input.rotationEnabled)
345
+ return input.eligible.find((account) => account.id === input.activeCredentialId) ?? null;
346
+ return pinned ?? (input.eligible.length ? input.eligible[subscriptionAccountShardIndex(input.sessionId, input.eligible.length)] : null);
347
+ }
348
+
349
+ // src/claude-subscription-capacity.ts
350
+ var UNKNOWN_RESET_RECHECK_MS = 6e4;
351
+ var allowed = (status2) => status2 === "allowed" || status2 === "allowed_warning";
352
+ function claudeSubscriptionCapacity(usage, upstreamModelId, now, options = {}) {
353
+ const available = { available: true, nextCheckAt: null, blockingWindows: [] };
354
+ if (!usage.connected || usage.refreshStatus === "reconnect")
355
+ return { ...available, available: false };
356
+ const relevant = (id) => {
357
+ if (id === "seven_day_opus") return /^claude-opus-/.test(upstreamModelId);
358
+ if (id === "seven_day_sonnet") return /^claude-sonnet-/.test(upstreamModelId);
359
+ return id === "five_hour" || id === "seven_day";
360
+ };
361
+ const applicable = usage.windows.filter((window) => relevant(window.id));
362
+ const relevantClaim = (id) => id === "seven_day_opus" || id === "seven_day_sonnet" ? relevant(id) : true;
363
+ const restrictions = (usage.requestRestrictions ?? []).filter(
364
+ (item) => item.status === "rejected" && (item.upstreamModelId === upstreamModelId || item.upstreamModelId === null && relevantClaim(item.representativeClaim)) && !applicable.some(
365
+ (window) => window.source === "provider" && window.usedPercent !== null && window.usedPercent < 100 && window.id === item.representativeClaim && window.observedAt > item.observedAt
366
+ ) && (item.resetsAt ? Date.parse(item.resetsAt) : Date.parse(item.observedAt) + UNKNOWN_RESET_RECHECK_MS) > now.getTime()
367
+ );
368
+ const blocking = applicable.filter((window) => {
369
+ if ((window.resetsAt ? Date.parse(window.resetsAt) : Date.parse(window.observedAt) + UNKNOWN_RESET_RECHECK_MS) <= now.getTime())
370
+ return false;
371
+ return window.status === "rejected" || window.status === null && (window.usedPercent ?? 0) >= 100;
372
+ });
373
+ const request = usage.requestStatus;
374
+ const exactModel = request?.upstreamModelId === upstreamModelId;
375
+ const relevantRequest = request && (exactModel || request.upstreamModelId === null && relevantClaim(request.representativeClaim));
376
+ const freshRequest = relevantRequest && applicable.every((window) => window.observedAt <= request.observedAt) && restrictions.every((item) => item.observedAt <= request.observedAt);
377
+ const planResetElapsed = request && (request.resetsAt ? Date.parse(request.resetsAt) : Date.parse(request.observedAt) + UNKNOWN_RESET_RECHECK_MS) <= now.getTime();
378
+ const exactAllowance = (usage.requestRestrictions ?? []).find(
379
+ (item) => item.upstreamModelId === upstreamModelId && allowed(item.status) && blocking.every((window) => window.observedAt < item.observedAt) && restrictions.every((restriction) => restriction.observedAt < item.observedAt)
380
+ );
381
+ if (exactAllowance) return available;
382
+ if (freshRequest && exactModel && allowed(request.status) && !planResetElapsed) return available;
383
+ const mayUseOverage = options.allowOverage !== false;
384
+ const overageResetElapsed = request?.overageResetsAt && Date.parse(request.overageResetsAt) <= now.getTime();
385
+ if (freshRequest && request.status === "rejected" && mayUseOverage && allowed(request.overageStatus) && !overageResetElapsed)
386
+ return available;
387
+ const rejectedRequest = freshRequest && request.status === "rejected" && !planResetElapsed;
388
+ if (blocking.length === 0 && !rejectedRequest && restrictions.length === 0) return available;
389
+ const resets = blocking.map(
390
+ (window) => window.resetsAt ? Date.parse(window.resetsAt) : Date.parse(window.observedAt) + UNKNOWN_RESET_RECHECK_MS
391
+ );
392
+ if (rejectedRequest)
393
+ resets.push(
394
+ request.resetsAt ? Date.parse(request.resetsAt) : Date.parse(request.observedAt) + UNKNOWN_RESET_RECHECK_MS
395
+ );
396
+ resets.push(
397
+ ...restrictions.map(
398
+ (item) => item.resetsAt ? Date.parse(item.resetsAt) : Date.parse(item.observedAt) + UNKNOWN_RESET_RECHECK_MS
399
+ )
400
+ );
401
+ let nextCheckAt = Math.max(...resets);
402
+ if (rejectedRequest && mayUseOverage && request.overageStatus === "rejected" && request.overageResetsAt) {
403
+ const overageReset = Date.parse(request.overageResetsAt);
404
+ if (overageReset <= now.getTime()) return available;
405
+ nextCheckAt = Math.min(nextCheckAt, overageReset);
406
+ }
407
+ return {
408
+ available: false,
409
+ nextCheckAt: new Date(nextCheckAt),
410
+ blockingWindows: blocking.map((window) => window.id)
411
+ };
412
+ }
413
+
414
+ // src/index.ts
46
415
  var envName = /^[A-Za-z_][A-Za-z0-9_]*$/;
47
416
  var registryId = /^[A-Za-z0-9_-]+$/;
48
417
  var DEFAULT_OPENROUTER_MODEL_ID = "openrouter/nvidia/nemotron-3-super-120b-a12b:free";
@@ -51,12 +420,13 @@ var DEFAULT_MODEL_COST_POLICY_JSON = JSON.stringify({
51
420
  });
52
421
  var SANDBOX_ARCHIVE_CAPTURE_MAX_TIMEOUT_MS = 60 * 6e4;
53
422
  var SANDBOX_ARCHIVE_CAPTURE_SETTLEMENT_GRACE_MS = 1e4;
423
+ var SANDBOX_DEADLINE_COMMAND_STOP_GRACE_MS = 12e4;
54
424
  var SANDBOX_SNAPSHOT_MAX_TIMEOUT_MS = SANDBOX_ARCHIVE_CAPTURE_MAX_TIMEOUT_MS - SANDBOX_ARCHIVE_CAPTURE_SETTLEMENT_GRACE_MS;
55
425
  var GOOGLE_DRIVE_PROVIDER_REQUEST_TIMEOUT_MAX_MS = 6e4;
56
426
  var GOOGLE_DRIVE_PROVIDER_RETRY_DELAY_MAX_MS = 6e4;
57
427
  var SANDBOX_LIFECYCLE_TRANSITION_MAX_WAIT_MS = 60 * 6e4;
58
428
  var SANDBOX_LIFECYCLE_RETRY_HANDOFF_GRACE_MS = 1e4;
59
- var EnvBoolean = z.preprocess((value) => {
429
+ var EnvBoolean = z2.preprocess((value) => {
60
430
  if (typeof value !== "string") {
61
431
  return value;
62
432
  }
@@ -68,10 +438,11 @@ var EnvBoolean = z.preprocess((value) => {
68
438
  return false;
69
439
  }
70
440
  return value;
71
- }, z.boolean());
441
+ }, z2.boolean());
442
+ var PublicHttpUrl = z2.string().url().max(2048).refine((value) => /^https?:\/\//iu.test(value), "must be an http(s) URL");
72
443
  var DEFAULT_GOAL_IDLE_BACKOFF_MS = [3e3, 3e4, 12e4, 3e5];
73
444
  var DEFAULT_GOAL_IDLE_BACKOFF_MAX_MS = 6e5;
74
- var EnvGoalIdleBackoffMs = z.preprocess((value) => {
445
+ var EnvGoalIdleBackoffMs = z2.preprocess((value) => {
75
446
  if (typeof value !== "string") return value;
76
447
  const source = value.trim();
77
448
  if (!source) return void 0;
@@ -79,8 +450,8 @@ var EnvGoalIdleBackoffMs = z.preprocess((value) => {
79
450
  const trimmed = entry.trim();
80
451
  return trimmed === "" ? Number.NaN : Number(trimmed);
81
452
  });
82
- }, z.array(z.number().int().nonnegative()).min(1, "OPENGENI_GOAL_IDLE_BACKOFF_MS must list at least one delay in milliseconds").readonly());
83
- var EnvFirstPartyMcpTools = z.preprocess(
453
+ }, z2.array(z2.number().int().nonnegative()).min(1, "OPENGENI_GOAL_IDLE_BACKOFF_MS must list at least one delay in milliseconds").readonly());
454
+ var EnvFirstPartyMcpTools = z2.preprocess(
84
455
  (value) => {
85
456
  if (typeof value !== "string") return value;
86
457
  const source = value.trim();
@@ -94,7 +465,7 @@ var EnvFirstPartyMcpTools = z.preprocess(
94
465
  }
95
466
  return source.split(",").map((entry) => entry.trim());
96
467
  },
97
- z.array(FirstPartyMcpToolName).superRefine((tools, context) => {
468
+ z2.array(FirstPartyMcpToolName).superRefine((tools, context) => {
98
469
  const seen = /* @__PURE__ */ new Set();
99
470
  for (const [index, tool] of tools.entries()) {
100
471
  if (seen.has(tool)) {
@@ -141,176 +512,153 @@ var sandboxPreparationProfiles = {
141
512
  };
142
513
  var AGENT_INSTRUCTIONS_CORE_PLACEHOLDER = "{{core}}";
143
514
  var DEFAULT_AGENT_INSTRUCTIONS = [
144
- "You are an OpenGeni workspace agent.",
145
- "Follow the user's task and any enabled pack or skill instructions for the current role.",
146
- "Work inside the sandbox workspace and use filesystem and shell tools when useful.",
515
+ "You are an OpenGeni workspace agent: a general assistant for questions, writing, research, analysis, and technical work.",
516
+ "Follow the user's task and the applicable Skill instructions for the current role.",
517
+ "When a task needs files or commands, work inside the sandbox workspace with the filesystem and shell tools.",
147
518
  "Repository resources are mounted under repos/<host>/<owner>/<repo> unless the session specifies another collision-free mount path.",
148
519
  "File resources are mounted under .opengeni/files/<file-id>/ unless the session specifies another mount path.",
149
520
  "Attached files are mounted read-only; copy them before modifying.",
150
521
  "Installed and selected Skills appear in the session Skill index; follow its reading instructions and any role-specific guidance.",
151
- "Use Checkov, Terraform, Azure CLI, git provider CLIs, and repository tools when relevant; gh, glab, and az repos are pre-authenticated when the host brokers matching git credentials.",
152
- "When the Azure sandbox preparation profile is enabled and service-principal variables are present, the sandbox is pre-authenticated with normal Azure CLI before work starts.",
153
- "Treat code-changing work as GitOps work: create a focused branch/commit/PR when git provider credentials are available; otherwise report exact commands and blockers.",
154
- "Return concise, factual summaries with files changed, commands run, and remaining blockers.",
522
+ "Provider CLIs such as gh, glab, and az may be pre-authenticated by the host through brokered git credentials or a sandbox preparation profile; try them before asking for credentials.",
523
+ "When the Git repository you change has a remote and git provider credentials are available, work on a focused branch and open a pull request.",
524
+ "Otherwise leave changes in the working tree and do not create or mention branches, commits, or pull requests unless the user asks; if the repository has a remote, say the changes are not pushed, and if the user asks for something you cannot make, say what blocks it.",
525
+ "Answer questions directly and briefly; after making changes, say what changed, how you checked it, and anything still blocked.",
155
526
  AGENT_INSTRUCTIONS_CORE_PLACEHOLDER
156
527
  ].join(" ");
157
- var McpServerConnectionRefSchema = z.object({
158
- // Standalone ids are UUIDs; embedded hosts may use any stable opaque id.
159
- connectionId: z.string().min(1).optional(),
160
- provider: z.string().min(1).max(128).optional(),
161
- providerDomain: z.string().min(1),
162
- kind: z.enum(["oauth2", "api_key", "app_install", "delegated"]).optional(),
163
- scopes: z.array(z.string().min(1)).optional(),
164
- resource: z.string().min(1).optional(),
165
- selectedResources: z.array(
166
- z.object({
167
- id: z.string().min(1).max(512),
168
- kind: z.literal("repository")
169
- }).strict()
170
- ).min(1).max(256).superRefine((resources, context) => {
171
- const seen = /* @__PURE__ */ new Set();
172
- for (const [index, resource] of resources.entries()) {
173
- const key = `${resource.kind}\0${resource.id}`;
174
- if (seen.has(key)) {
175
- context.addIssue({
176
- code: "custom",
177
- message: "selectedResources must not contain duplicates",
178
- path: [index]
179
- });
180
- }
181
- seen.add(key);
528
+ var McpServerConnectionRefSchema = PublicMcpServerConnectionRef;
529
+ var McpOperationRecoverySchema = z2.record(
530
+ z2.string().min(1).max(512).regex(/^\S+$/u),
531
+ z2.object({ observerTool: z2.string().min(1).max(512).regex(/^\S+$/u) }).strict()
532
+ ).superRefine((bindings, context) => {
533
+ for (const [originalTool, binding] of Object.entries(bindings)) {
534
+ if (originalTool === binding.observerTool) {
535
+ context.addIssue({
536
+ code: "custom",
537
+ path: [originalTool, "observerTool"],
538
+ message: "MCP recovery observer must differ from the original tool"
539
+ });
182
540
  }
183
- }).optional(),
184
- authoritySource: z.literal("host").optional(),
185
- subjectScope: z.enum(["workspace", "subject"]).optional()
186
- }).strict().superRefine((reference, context) => {
187
- if (reference.authoritySource === "host" && !reference.connectionId) {
188
- context.addIssue({
189
- code: "custom",
190
- message: "host authority requires connectionId",
191
- path: ["connectionId"]
192
- });
193
- }
194
- if (!reference.selectedResources) return;
195
- if (!reference.connectionId) {
196
- context.addIssue({
197
- code: "custom",
198
- message: "selectedResources requires connectionId",
199
- path: ["connectionId"]
200
- });
201
- }
202
- if (!reference.provider) {
203
- context.addIssue({
204
- code: "custom",
205
- message: "selectedResources requires provider",
206
- path: ["provider"]
207
- });
208
541
  }
209
542
  });
210
- var DEFAULT_MODAL_IMAGE_REF = "opengenipublicneuacr.azurecr.io/opengeni-desktop@sha256:c3bd17b8841de1bff9bb2777aad422cf8c75de78e2c30f9ac81d0cd6810a1b78";
211
- var SettingsSchema = z.object({
212
- serviceName: z.string().default("opengeni"),
213
- environment: z.string().default("local"),
214
- deploymentRevision: z.string().default("dev"),
543
+ var DEFAULT_MODAL_IMAGE_REF = "opengenipublicneuacr.azurecr.io/opengeni-desktop@sha256:554d5b324a580071dd669e7c1190931e498eb1b397412a27c7de84a430ef393a";
544
+ var SettingsSchema = z2.object({
545
+ serviceName: z2.string().default("opengeni"),
546
+ environment: z2.string().default("local"),
547
+ deploymentRevision: z2.string().default("dev"),
215
548
  // The release-train version baked into official images (OPENGENI_SERVER_VERSION).
216
549
  // Absent on dev/source builds — consumers must treat it as optional.
217
- serverVersion: z.string().optional(),
218
- databaseUrl: z.string().default("postgres://opengeni:opengeni@127.0.0.1:5432/opengeni"),
550
+ serverVersion: z2.string().optional(),
551
+ databaseUrl: z2.string().default("postgres://opengeni:opengeni@127.0.0.1:5432/opengeni"),
552
+ // API application pool only. Managed auth owns its separate pool; readiness
553
+ // reuses the application handle. Keep the historical default for existing
554
+ // deployments and let operators budget this per process with replica counts.
555
+ apiDatabasePoolMax: z2.coerce.number().int().positive().default(32),
219
556
  // Step I (§7.8 runtime half). Dedicated Postgres schema for the EMBEDDED
220
557
  // topology. Default "" → standalone: no search_path scoping, server default
221
558
  // (`public`). When set (e.g. "opengeni"), the db handle + the managed-auth
222
559
  // pool send `search_path = "<dbSchema>","opengeni_private","public"` so every
223
560
  // query resolves into the dedicated schema with NO query rewrite (schema-isolation contract F1).
224
- dbSchema: z.string().default(""),
561
+ dbSchema: z2.string().default(""),
225
562
  // Step I (§7.7). RLS posture. "force" (default) = today's FORCE-RLS via the
226
563
  // non-owner `opengeni_app` role. "scoped" = the embedded owner-role path (the
227
564
  // GUC is still emitted defensively, so the query path is identical).
228
- rlsStrategy: z.enum(["force", "scoped"]).default("force"),
565
+ rlsStrategy: z2.enum(["force", "scoped"]).default("force"),
229
566
  // Exact PostgreSQL login identity required by the standalone FORCE-RLS
230
567
  // startup/readiness assertion. Embedded `scoped` hosts own their role model
231
568
  // and are deliberately not constrained to this name.
232
- runtimeDatabaseRole: z.string().min(1).default("opengeni_app"),
233
- natsUrl: z.string().default("nats://127.0.0.1:4222"),
234
- temporalHost: z.string().default("127.0.0.1:7233"),
235
- temporalNamespace: z.string().default("default"),
236
- temporalTaskQueue: z.string().default("opengeni-runs-ts"),
569
+ runtimeDatabaseRole: z2.string().min(1).default("opengeni_app"),
570
+ natsUrl: z2.string().default("nats://127.0.0.1:4222"),
571
+ temporalHost: z2.string().default("127.0.0.1:7233"),
572
+ temporalNamespace: z2.string().default("default"),
573
+ temporalTaskQueue: z2.string().default("opengeni-runs-ts"),
237
574
  temporalTlsEnabled: EnvBoolean.default(false),
238
- temporalApiKey: z.string().optional(),
239
- temporalTlsServerName: z.string().optional(),
240
- temporalTlsRootCaCertificateBase64: z.string().optional(),
241
- temporalTlsClientCertificateBase64: z.string().optional(),
242
- temporalTlsClientPrivateKeyBase64: z.string().optional(),
243
- startupDependencyRetryAttempts: z.coerce.number().int().positive().default(30),
244
- startupDependencyRetryInitialDelayMs: z.coerce.number().int().positive().default(1e3),
245
- startupDependencyRetryMaxDelayMs: z.coerce.number().int().positive().default(5e3),
246
- turnWorkerConcurrencyMode: z.enum(["fixed", "resource-based"]).default("fixed"),
247
- turnWorkerMaxConcurrentTurns: z.coerce.number().int().positive().max(2e3).default(16),
248
- turnWorkerTargetCpuUsage: z.coerce.number().positive().max(1).default(0.8),
249
- turnWorkerTargetMemoryUsage: z.coerce.number().positive().max(0.8).default(0.75),
575
+ temporalApiKey: z2.string().optional(),
576
+ temporalTlsServerName: z2.string().optional(),
577
+ temporalTlsRootCaCertificateBase64: z2.string().optional(),
578
+ temporalTlsClientCertificateBase64: z2.string().optional(),
579
+ temporalTlsClientPrivateKeyBase64: z2.string().optional(),
580
+ startupDependencyRetryAttempts: z2.coerce.number().int().positive().default(30),
581
+ startupDependencyRetryInitialDelayMs: z2.coerce.number().int().positive().default(1e3),
582
+ startupDependencyRetryMaxDelayMs: z2.coerce.number().int().positive().default(5e3),
583
+ turnWorkerConcurrencyMode: z2.enum(["fixed", "resource-based"]).default("fixed"),
584
+ turnWorkerMaxConcurrentTurns: z2.coerce.number().int().positive().max(2e3).default(16),
585
+ turnWorkerTargetCpuUsage: z2.coerce.number().positive().max(1).default(0.8),
586
+ turnWorkerTargetMemoryUsage: z2.coerce.number().positive().max(0.8).default(0.75),
250
587
  // Admission and emergency recovery are deliberately separate control loops.
251
588
  // The Temporal tuner stops polling at the lower target; only genuine danger
252
589
  // may invoke the disruptive graceful-drain fallback.
253
- turnWorkerEmergencyMemoryUsage: z.coerce.number().min(0.85).max(0.95).default(0.9),
254
- turnWorkerMemoryGuardIntervalMs: z.coerce.number().int().min(1e3).max(6e4).default(5e3),
255
- turnWorkerMemoryGuardSustainMs: z.coerce.number().int().min(5e3).max(3e5).default(3e4),
590
+ turnWorkerEmergencyMemoryUsage: z2.coerce.number().min(0.85).max(0.95).default(0.9),
591
+ turnWorkerMemoryGuardIntervalMs: z2.coerce.number().int().min(1e3).max(6e4).default(5e3),
592
+ turnWorkerMemoryGuardSustainMs: z2.coerce.number().int().min(5e3).max(3e5).default(3e4),
256
593
  observabilityStructuredLogs: EnvBoolean.default(false),
257
594
  observabilityMetricsEnabled: EnvBoolean.default(true),
258
- observabilityOtlpEndpoint: z.string().url().optional(),
259
- observabilityOtlpHeaders: z.string().default(""),
595
+ observabilityOtlpEndpoint: z2.string().url().optional(),
596
+ observabilityOtlpHeaders: z2.string().default(""),
597
+ observabilityDiagnosticsEndpoint: z2.string().url().optional(),
598
+ observabilityDiagnosticsHeaders: z2.string().default(""),
260
599
  analyticsEnabled: EnvBoolean.default(false),
261
600
  analyticsConsentRequired: EnvBoolean.default(true),
262
- analyticsReoClientId: z.string().max(128).regex(/^[A-Za-z0-9_-]+$/u).optional(),
263
- analyticsPosthogProjectKey: z.string().min(1).max(256).optional(),
264
- analyticsPosthogHost: z.string().url().max(2048).optional(),
265
- analyticsGa4MeasurementId: z.string().max(32).regex(/^G-[A-Z0-9]+$/u).optional(),
266
- publicBaseUrl: z.string().url().optional(),
601
+ analyticsReoClientId: z2.string().max(128).regex(/^[A-Za-z0-9_-]+$/u).optional(),
602
+ analyticsPosthogProjectKey: z2.string().min(1).max(256).optional(),
603
+ analyticsPosthogHost: z2.string().url().max(2048).optional(),
604
+ analyticsGa4MeasurementId: z2.string().max(32).regex(/^G-[A-Z0-9]+$/u).optional(),
605
+ // Optional operator-owned legal documents linked from the signed-out console.
606
+ // Unset by default, so a self-hosted deployment never shows another operator's
607
+ // policies; the managed service points these at its own published pages.
608
+ legalPrivacyPolicyUrl: PublicHttpUrl.optional(),
609
+ legalTermsOfServiceUrl: PublicHttpUrl.optional(),
610
+ // Optional operator support address, shown as a "Contact support" mailto link
611
+ // on the signed-out page and in the Help menu. Unset by default for the same
612
+ // reason as the legal links.
613
+ supportEmail: z2.string().trim().email().max(254).optional(),
614
+ publicBaseUrl: z2.string().url().optional(),
615
+ // Product documentation the web console links from its Help menu. Absent
616
+ // means the public OpenGeni docs; `none` hides the link for deployments that
617
+ // publish no documentation of their own.
618
+ documentationUrl: z2.preprocess(
619
+ (value) => typeof value === "string" ? value.trim().toLowerCase() === "none" ? null : value.trim() : value,
620
+ z2.url({ protocol: /^https?$/u, error: "must be an absolute http(s) URL or none" }).nullable().default(DEFAULT_OPENGENI_DOCUMENTATION_URL)
621
+ ),
267
622
  // Standards-based OAuth authorization server for external workspace MCP
268
623
  // clients. Opt-in because it creates a new public authentication surface.
269
624
  mcpOauthEnabled: EnvBoolean.default(false),
270
- // Forwarded client addresses are ignored by default. Operators may trust an
625
+ // Client source address for every API rate limit, abuse quota, and auth
626
+ // session record (Better Auth, MCP OAuth registration, enrollment, account
627
+ // setup, login transactions). Forwarded client addresses are ignored by
628
+ // default so a caller cannot choose its own bucket. Operators may trust an
271
629
  // exact number of proxy hops only when direct access to the API is blocked.
272
- mcpOauthTrustedProxyHops: z.coerce.number().int().min(0).max(16).default(0),
630
+ apiTrustedProxyHops: z2.coerce.number().int().min(0).max(16).default(0),
631
+ // Optional comma-separated CIDRs (or single addresses) of the proxy that
632
+ // connects to the API. When set, forwarded client addresses are honored
633
+ // only for requests whose transport peer is inside one of them.
634
+ apiTrustedProxyCidrs: z2.string().default(""),
273
635
  // Browser origin when the web app and API use separate origins in local
274
636
  // development. Production normally leaves this unset and uses publicBaseUrl.
275
- webBaseUrl: z.string().url().optional(),
637
+ webBaseUrl: z2.string().url().optional(),
276
638
  // Base URL for the bring-your-own-compute agent release assets the get.<domain>
277
639
  // install routes redirect to. Defaults to this repo's GitHub Releases. The route
278
640
  // appends `/download/agent-v<ver>/<asset>`.
279
- agentReleasesBaseUrl: z.string().url().default("https://github.com/Cloudgeni-ai/opengeni/releases"),
641
+ agentReleasesBaseUrl: z2.string().url().default("https://github.com/Cloudgeni-ai/opengeni/releases"),
280
642
  // Explicit operator-controlled promotion pointer for `/agent/latest/*`.
281
643
  // Versioned agent releases are immutable; changing this setting promotes or
282
644
  // rolls back the stable channel without moving or deleting a provider tag.
283
- agentStableVersion: z.string().regex(/^(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)$/u).default("0.1.16"),
645
+ agentStableVersion: z2.string().regex(/^(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)$/u).default("0.1.16"),
284
646
  // Optional independent beta-channel pointer. When unset, the beta update
285
647
  // manifest route is unavailable rather than silently serving stable.
286
- agentBetaVersion: z.string().regex(/^(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)$/u).optional(),
648
+ agentBetaVersion: z2.string().regex(/^(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)$/u).optional(),
287
649
  productAccessMode: ProductAccessMode.default("local"),
288
- // --- canonical organization-tenancy authority activation, default OFF ---
289
- // The named PRE-ACTIVATION opt-out for the organization-tenancy program. FALSE (the
290
- // default, and the value an operator leaves in place to decline or defer) means
291
- // this deployment stays on the reversible legacy workspace-owned lane: no phase-F
292
- // subsystem may switch its access decision to organization/membership authority
293
- // ids. TRUE is an operator's explicit statement that the activation preconditions
294
- // in docs/organization-tenancy.md have been proven for this deployment and that
295
- // the one-way boundary is accepted.
296
- //
297
- // This is NOT a kill switch and NOT a rollback: once an activation migration has
298
- // committed, setting it back to false does not restore the legacy authority - only
299
- // forward recovery is available. It also grants and revokes nothing by itself;
300
- // every individual authorization decision keeps its own fences.
301
- //
302
- // No runtime path reads it yet: canonical activation (phase F) is unshipped, so
303
- // the flag exists to reserve the name, pin the safe default, and give every future
304
- // activation slice one gate to consult. EnvBoolean (NOT z.coerce.boolean(), which
305
- // coerces "false" -> true and would activate the moment an operator wrote the
306
- // variable out to disable it).
307
- organizationTenancyCanonicalActivationEnabled: EnvBoolean.default(false),
308
650
  billingMode: BillingMode.default("disabled"),
651
+ // Explicit launch gate for the one-time $10 verified self-service signup grant.
652
+ // A migration or deployment alone must not start issuing live credits.
653
+ verifiedSignupTrialCreditsEnabled: EnvBoolean.default(false),
309
654
  entitlementsMode: EntitlementsMode.default("none"),
310
655
  usageLimitsMode: UsageLimitsMode.default("none"),
311
- staticEntitlementsJson: z.string().default("{}"),
312
- staticUsageLimitsJson: z.string().default("{}"),
313
- delegationSecret: z.string().optional(),
656
+ // Gate new allowance policies until every API/worker in the fleet enforces
657
+ // them. Persisted-policy enforcement and recovery reads/clears stay active.
658
+ usageAllowancesEnabled: EnvBoolean.default(false),
659
+ staticEntitlementsJson: z2.string().default("{}"),
660
+ staticUsageLimitsJson: z2.string().default("{}"),
661
+ delegationSecret: z2.string().optional(),
314
662
  defaultFirstPartyMcpTools: EnvFirstPartyMcpTools,
315
663
  allowedFirstPartyMcpTools: EnvFirstPartyMcpTools,
316
664
  // sandbox workspace scoped stream-token HMAC secret (sandbox contract §C.3 / stream-token availability contract).
@@ -318,7 +666,7 @@ var SettingsSchema = z.object({
318
666
  // family, `ogs_` vs `ogd_` prefix). REQUIRED-WHEN-DESKTOP, but the absence of
319
667
  // BOTH while sandboxDesktopEnabled=true is a GRACEFUL DEGRADE (DesktopStream
320
668
  // transport:null + a loud boot warning), NOT a hard boot-fail (stream-token availability contract).
321
- streamTokenSecret: z.string().optional(),
669
+ streamTokenSecret: z2.string().optional(),
322
670
  // The desktop input plane (raw stream:control writes) is OFF in v1: even a
323
671
  // holder of stream:control gets 403 until this flips. Keeps stream:control a
324
672
  // declared-but-inert permission so later hardening is a flag flip.
@@ -341,47 +689,51 @@ var SettingsSchema = z.object({
341
689
  // Optional release-coherent bootstrap hint for custom rigs/connected machines
342
690
  // that do not carry the stock-image ogtool binary. Exact stable versions only:
343
691
  // the agent must never guess a tag or silently install `latest`.
344
- ogtoolPackageSpec: z.string().regex(/^@opengeni\/ogtool@(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)$/u).optional(),
345
- environmentsEncryptionKey: z.string().optional(),
692
+ ogtoolPackageSpec: z2.string().regex(/^@opengeni\/ogtool@(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)\.(?:0|[1-9]\d*)$/u).optional(),
693
+ environmentsEncryptionKey: z2.string().optional(),
346
694
  integrationsEnabled: EnvBoolean.default(false),
347
- integrationsStateSecret: z.string().optional(),
695
+ integrationsStateSecret: z2.string().optional(),
696
+ integrationsOauthShortStateEnabled: EnvBoolean.default(false),
348
697
  integrationsAllowPrivateNetworkTargets: EnvBoolean.default(false),
349
- integrationsOauthClientsJson: z.string().default("{}"),
350
- slackClientId: z.string().optional(),
351
- slackClientSecret: z.string().optional(),
352
- slackSigningSecret: z.string().optional(),
698
+ integrationsOauthClientsJson: z2.string().default("{}"),
699
+ gmailWatchTopicName: z2.string().regex(/^projects\/[^/]+\/topics\/[^/]+$/u).optional(),
700
+ slackClientId: z2.string().optional(),
701
+ slackClientSecret: z2.string().optional(),
702
+ slackSigningSecret: z2.string().optional(),
703
+ // Unlisted apps share Slack’s restricted history/replies quota across all tokens.
704
+ slackAccessMode: z2.enum(["limited", "full"]).default("limited"),
353
705
  slackBotDisplayName: OpenGeniSlackBotDisplayName.default("OpenGeni"),
354
- slackCommand: z.string().trim().regex(/^\/[a-z0-9_-]{1,31}$/u).default("/opengeni"),
355
- googleDriveClientId: z.string().optional(),
356
- googleDriveClientSecret: z.string().optional(),
357
- googleDriveSyncMaxItems: z.coerce.number().int().positive().max(1e4).default(500),
358
- googleDriveSyncMaxBytes: z.coerce.number().int().positive().max(5e9).default(5e8),
359
- googleDriveSyncMaxFileBytes: z.coerce.number().int().positive().max(5e9).default(1e8),
360
- googleDriveSyncMaxProviderRequests: z.coerce.number().int().positive().max(1e4).default(1e3),
361
- googleDriveSyncMaxElapsedSeconds: z.coerce.number().int().positive().max(3600).default(300),
362
- googleDriveSyncMaxFailureDetails: z.coerce.number().int().positive().max(100).default(25),
363
- googleDriveProviderRequestTimeoutMs: z.coerce.number().int().min(1e3).max(GOOGLE_DRIVE_PROVIDER_REQUEST_TIMEOUT_MAX_MS).default(3e4),
364
- googleDriveProviderRetryAttempts: z.coerce.number().int().min(1).max(5).default(3),
365
- googleDriveProviderRetryInitialDelayMs: z.coerce.number().int().positive().max(3e4).default(250),
366
- googleDriveProviderRetryMaxDelayMs: z.coerce.number().int().positive().max(GOOGLE_DRIVE_PROVIDER_RETRY_DELAY_MAX_MS).default(5e3),
367
- googleDriveProviderRetryBudgetMs: z.coerce.number().int().positive().max(12e4).default(15e3),
368
- fikenClientId: z.string().optional(),
369
- fikenClientSecret: z.string().optional(),
706
+ slackCommand: z2.string().trim().regex(/^\/[a-z0-9_-]{1,31}$/u).default("/opengeni"),
707
+ googleDriveClientId: z2.string().optional(),
708
+ googleDriveClientSecret: z2.string().optional(),
709
+ googleDriveSyncMaxItems: z2.coerce.number().int().positive().max(1e4).default(500),
710
+ googleDriveSyncMaxBytes: z2.coerce.number().int().positive().max(5e9).default(5e8),
711
+ googleDriveSyncMaxFileBytes: z2.coerce.number().int().positive().max(5e9).default(1e8),
712
+ googleDriveSyncMaxProviderRequests: z2.coerce.number().int().positive().max(1e4).default(1e3),
713
+ googleDriveSyncMaxElapsedSeconds: z2.coerce.number().int().positive().max(3600).default(300),
714
+ googleDriveSyncMaxFailureDetails: z2.coerce.number().int().positive().max(100).default(25),
715
+ googleDriveProviderRequestTimeoutMs: z2.coerce.number().int().min(1e3).max(GOOGLE_DRIVE_PROVIDER_REQUEST_TIMEOUT_MAX_MS).default(3e4),
716
+ googleDriveProviderRetryAttempts: z2.coerce.number().int().min(1).max(5).default(3),
717
+ googleDriveProviderRetryInitialDelayMs: z2.coerce.number().int().positive().max(3e4).default(250),
718
+ googleDriveProviderRetryMaxDelayMs: z2.coerce.number().int().positive().max(GOOGLE_DRIVE_PROVIDER_RETRY_DELAY_MAX_MS).default(5e3),
719
+ googleDriveProviderRetryBudgetMs: z2.coerce.number().int().positive().max(12e4).default(15e3),
720
+ fikenClientId: z2.string().optional(),
721
+ fikenClientSecret: z2.string().optional(),
370
722
  googleDriveWorkspaceEventsEnabled: EnvBoolean.optional(),
371
- atlassianClientId: z.string().optional(),
372
- atlassianClientSecret: z.string().optional(),
723
+ atlassianClientId: z2.string().optional(),
724
+ atlassianClientSecret: z2.string().optional(),
373
725
  // Undefined is meaningful: the migration boundary persists the product
374
726
  // default of 3 when no deployment override is supplied.
375
- maxNestedAgentDepth: z.coerce.number().int().nonnegative().max(MAX_NESTED_AGENT_DEPTH).optional(),
727
+ maxNestedAgentDepth: z2.coerce.number().int().nonnegative().max(MAX_NESTED_AGENT_DEPTH).optional(),
376
728
  // Operator OAuth apps for first-party social connectors, keyed by provider
377
729
  // id ("x", "reddit"): {"x":{"clientId":"...","clientSecret":"..."}}.
378
- socialOauthClientsJson: z.string().default("{}"),
730
+ socialOauthClientsJson: z2.string().default("{}"),
379
731
  // Session goal guard rails. Goals are designed for runs that legitimately
380
732
  // span days, so length is bounded by explicit completion/pause and budget
381
733
  // exhaustion, never by count. goalMaxAutoContinuations is therefore UNSET
382
734
  // by default (no cap); deployments may configure one, and it then acts as a
383
735
  // hard ceiling that per-goal overrides can only lower.
384
- goalMaxAutoContinuations: z.coerce.number().int().positive().optional(),
736
+ goalMaxAutoContinuations: z2.coerce.number().int().positive().optional(),
385
737
  // Idle backoff between CONSECUTIVE no-input goal continuations. This is
386
738
  // pacing, not a cap: the first continuation after a turn that consumed any
387
739
  // external input is immediate, the n-th consecutive no-input continuation
@@ -389,7 +741,7 @@ var SettingsSchema = z.object({
389
741
  // any new input (machine input, human/API prompt, Steer) wakes the session
390
742
  // immediately. The delay never exceeds goalIdleBackoffMaxMs.
391
743
  goalIdleBackoffMs: EnvGoalIdleBackoffMs.default(DEFAULT_GOAL_IDLE_BACKOFF_MS),
392
- goalIdleBackoffMaxMs: z.coerce.number().int().positive().default(DEFAULT_GOAL_IDLE_BACKOFF_MAX_MS),
744
+ goalIdleBackoffMaxMs: z2.coerce.number().int().positive().default(DEFAULT_GOAL_IDLE_BACKOFF_MAX_MS),
393
745
  // Child lifecycle notices: a child session's requires_action freeze, its
394
746
  // resolution, a direct Pause, a provider-capacity wait, and goal progress
395
747
  // become typed `session_system_updates` rows for the parent (in addition to
@@ -403,13 +755,6 @@ var SettingsSchema = z.object({
403
755
  // into @opengeni/db once at boot.
404
756
  // Env: OPENGENI_CHILD_LIFECYCLE_NOTICES_ENABLED.
405
757
  childLifecycleNoticesEnabled: EnvBoolean.default(false),
406
- // Explicit host-owned MCP connection authority is a rolling protocol
407
- // activation. Keep it off while any API, worker, or browser bundle predates
408
- // the authority discriminator; enable it only after the whole fleet runs an
409
- // image that understands host refs. Legacy markerless non-UUID refs remain a
410
- // separate compatibility lane for already-persisted embedding integrations.
411
- // Env: OPENGENI_HOST_MCP_AUTHORITY_SOURCE_ADMISSION_ENABLED.
412
- hostMcpAuthoritySourceAdmissionEnabled: EnvBoolean.default(false),
413
758
  // Per-channel and per-DM Slack workspace routing. Default ON. A channel does
414
759
  // not count a personal workspace as a candidate, so an organization with one
415
760
  // shared workspace resolves it as the sole candidate and never asks; the
@@ -424,17 +769,17 @@ var SettingsSchema = z.object({
424
769
  // session turn. Effectively unbounded by default for the same reason as
425
770
  // above; the graceful max-turns valve (idle + goal continuation, never a
426
771
  // session failure) remains as inert safety should a deployment set a cap.
427
- agentMaxModelCallsPerTurn: z.coerce.number().int().positive().default(1e6),
772
+ agentMaxModelCallsPerTurn: z2.coerce.number().int().positive().default(1e6),
428
773
  // Deployment fallback for models that do not declare their own window.
429
774
  // Built-in billed GPT-5.6 Sol/Terra/Luna pin Codex's 272k catalog instead.
430
775
  // OpenGeni always performs one durable, portable plaintext compaction
431
776
  // transition; there is no provider/server/off mode ladder.
432
- contextWindowTokens: z.coerce.number().int().positive().default(105e4),
777
+ contextWindowTokens: z2.coerce.number().int().positive().default(105e4),
433
778
  // Optional model-catalog effective input ceiling. Codex and billed GPT-5.6
434
779
  // models expose this as raw context_window * effective_context_window_percent;
435
780
  // when absent, retain the deployment-level window-minus-reserved-output
436
781
  // behavior.
437
- contextEffectiveWindowTokens: z.coerce.number().int().positive().optional(),
782
+ contextEffectiveWindowTokens: z2.coerce.number().int().positive().optional(),
438
783
  // Proactive compaction threshold as a ratio of the model context window.
439
784
  // Defaults to 90%: compact as late as possible — retained context beats early
440
785
  // headroom now that per-model windows are declared honestly (input-effective,
@@ -442,7 +787,7 @@ var SettingsSchema = z.object({
442
787
  // absorbs any overshoot as one retried call rather than a dead session.
443
788
  // Clamped to [0.3, 0.9] so deployments can tune the trigger without
444
789
  // accidentally disabling compaction.
445
- contextCompactionThresholdRatio: z.coerce.number().default(0.9).transform((value) => {
790
+ contextCompactionThresholdRatio: z2.coerce.number().default(0.9).transform((value) => {
446
791
  if (!Number.isFinite(value)) {
447
792
  return 0.9;
448
793
  }
@@ -450,118 +795,127 @@ var SettingsSchema = z.object({
450
795
  }),
451
796
  // Tokens reserved for model output; subtracted from the window to get the
452
797
  // usable input budget B = contextWindowTokens - contextReservedOutputTokens.
453
- contextReservedOutputTokens: z.coerce.number().int().nonnegative().default(128e3),
798
+ contextReservedOutputTokens: z2.coerce.number().int().nonnegative().default(128e3),
454
799
  // Model-catalog auto-compact limit. When present it is clamped to
455
800
  // 90% of the raw window, matching Codex core's auto_compact_token_limit().
456
- contextAutoCompactThresholdTokens: z.coerce.number().int().positive().optional(),
801
+ contextAutoCompactThresholdTokens: z2.coerce.number().int().positive().optional(),
457
802
  // Provider-neutral fallback for canonical model-facing tool-result text.
458
803
  // The current stable Codex catalog policy is 10k tokens; the truncator adds
459
804
  // Codex's 1.2x JSON serialization allowance when applying it.
460
- modelToolOutputTruncationTokens: z.coerce.number().int().positive().default(1e4),
805
+ modelToolOutputTruncationTokens: z2.coerce.number().int().positive().default(1e4),
461
806
  authRequired: EnvBoolean.default(false),
462
- accessKey: z.string().optional(),
807
+ accessKey: z2.string().optional(),
463
808
  authAllowHealth: EnvBoolean.default(true),
464
809
  authAllowMetrics: EnvBoolean.default(false),
465
- apiHost: z.string().default("0.0.0.0"),
466
- apiPort: z.coerce.number().int().positive().default(8e3),
467
- workerHttpPort: z.coerce.number().int().positive().default(8001),
810
+ apiHost: z2.string().default("0.0.0.0"),
811
+ apiPort: z2.coerce.number().int().positive().default(8e3),
812
+ // When set, GET /metrics is served only on this dedicated internal listener
813
+ // and never on the public API port that an ingress forwards to.
814
+ apiMetricsPort: z2.coerce.number().int().positive().optional(),
815
+ workerHttpPort: z2.coerce.number().int().positive().default(8001),
468
816
  // Worker-side first-party MCP traffic stays on the deployment's internal
469
817
  // network. OPENGENI_MCP_URL remains the sandbox/external route used by
470
818
  // Codemode and remote placements.
471
- opengeniMcpInternalUrl: z.string().url().optional(),
472
- opengeniMcpUrl: z.string().url().optional(),
819
+ opengeniMcpInternalUrl: z2.string().url().optional(),
820
+ opengeniMcpUrl: z2.string().url().optional(),
473
821
  // Origins allowed to send browser cookies cross-origin. Other origins may
474
822
  // call the public API with bearer credentials, but never receive credentialed
475
823
  // CORS responses.
476
- corsAllowOriginRegex: z.string().default(String.raw`^https?://(localhost|127\.0\.0\.1)(:\d+)?$`),
477
- openaiProvider: z.enum(["openai", "azure"]).default("openai"),
478
- openaiApiKey: z.string().optional(),
479
- openaiBaseUrl: z.string().optional(),
480
- openaiModel: z.string().default("gpt-5.6-sol"),
481
- openaiAllowedModels: z.string().default("gpt-5.6-sol,gpt-5.6-terra,gpt-5.6-luna"),
824
+ corsAllowOriginRegex: z2.string().default(String.raw`^https?://(localhost|127\.0\.0\.1)(:\d+)?$`),
825
+ // Local development only (`local` access mode in the `local` environment):
826
+ // extra exact browser origins, comma-separated, that may call the
827
+ // unauthenticated API besides this stack's own web origin. The local browser
828
+ // boundary ignores `corsAllowOriginRegex`.
829
+ localAllowedOrigins: z2.string().optional(),
830
+ openaiProvider: z2.enum(["openai", "azure"]).default("openai"),
831
+ openaiApiKey: z2.string().optional(),
832
+ openaiBaseUrl: z2.string().optional(),
833
+ openaiModel: z2.string().default("gpt-6-astra"),
834
+ openaiAllowedModels: z2.string().default("gpt-6-astra,gpt-6-sol,gpt-6-luna"),
482
835
  // OpenGeni-managed Vercel AI Gateway. When configured, the two reviewed
483
836
  // Gateway models below are added to the managed-credit catalog. Workspace
484
837
  // Gateway keys use the encrypted connection broker and never this secret.
485
- vercelAiGatewayApiKey: z.string().optional(),
838
+ vercelAiGatewayApiKey: z2.string().optional(),
486
839
  /** Image adapter route; native hosted providers ignore this model. */
487
- imageGenerationModel: z.string().trim().min(1).max(256).default("openai/gpt-image-2"),
840
+ imageGenerationModel: z2.string().trim().min(1).max(256).default("openai/gpt-image-2"),
488
841
  /** Durable video generation uses the workspace-owned Gateway credential. */
489
- videoGenerationPollIntervalMs: z.coerce.number().int().min(1e3).max(6e4).default(5e3),
490
- videoGenerationRecoveryDeadlineMs: z.coerce.number().int().min(6e4).max(24 * 60 * 6e4).default(2 * 60 * 6e4),
491
- videoGenerationReferenceUrlTtlSeconds: z.coerce.number().int().min(300).max(6 * 60 * 60).default(60 * 60),
492
- videoGenerationMaxConcurrentPerWorkspace: z.coerce.number().int().min(1).max(16).default(2),
493
- videoGenerationWorkspaceQuotaBytes: z.coerce.number().int().positive().max(Number.MAX_SAFE_INTEGER).default(20 * 1024 * 1024 * 1024),
494
- videoGenerationTempDirectory: z.string().trim().min(1).max(1024).default("/tmp/opengeni-video"),
495
- videoGenerationFfprobePath: z.string().trim().min(1).max(1024).default("ffprobe"),
842
+ videoGenerationPollIntervalMs: z2.coerce.number().int().min(1e3).max(6e4).default(5e3),
843
+ videoGenerationRecoveryDeadlineMs: z2.coerce.number().int().min(6e4).max(24 * 60 * 6e4).default(2 * 60 * 6e4),
844
+ videoGenerationReferenceUrlTtlSeconds: z2.coerce.number().int().min(300).max(6 * 60 * 60).default(60 * 60),
845
+ videoGenerationMaxConcurrentPerWorkspace: z2.coerce.number().int().min(1).max(16).default(2),
846
+ videoGenerationWorkspaceQuotaBytes: z2.coerce.number().int().positive().max(Number.MAX_SAFE_INTEGER).default(20 * 1024 * 1024 * 1024),
847
+ videoGenerationTempDirectory: z2.string().trim().min(1).max(1024).default("/tmp/opengeni-video"),
848
+ videoGenerationFfprobePath: z2.string().trim().min(1).max(1024).default("ffprobe"),
496
849
  // OpenGeni's customer price, not a claim about the provider's delayed cost report.
497
850
  // The durable operation freezes the exact resulting price before provider submit.
498
- videoGenerationCredit480pMicrosPerSecond: z.coerce.number().int().positive().max(1e7).default(155e3),
499
- videoGenerationCredit720pMicrosPerSecond: z.coerce.number().int().positive().max(1e7).default(35e4),
851
+ videoGenerationCredit480pMicrosPerSecond: z2.coerce.number().int().positive().max(1e7).default(155e3),
852
+ videoGenerationCredit720pMicrosPerSecond: z2.coerce.number().int().positive().max(1e7).default(35e4),
500
853
  // Native composer voice input (browser MediaRecorder → API transcription).
501
854
  // Provider credentials stay server-side; ClientConfig only projects availability
502
855
  // and hard ceilings. Selection happens once before audio is sent — never retry
503
856
  // the same clip across vendors after an upstream request may have started.
504
- voiceInputMaxDurationSeconds: z.coerce.number().int().positive().max(600).default(60),
505
- voiceInputMaxSizeBytes: z.coerce.number().int().positive().max(25 * 1024 * 1024).default(25 * 1024 * 1024),
857
+ voiceInputMaxDurationSeconds: z2.coerce.number().int().positive().max(600).default(60),
858
+ voiceInputMaxSizeBytes: z2.coerce.number().int().positive().max(25 * 1024 * 1024).default(25 * 1024 * 1024),
506
859
  // Durable long-form capture uploads bounded chunks to object storage, then
507
860
  // normalizes them into provider-safe segments. It is advertised only when
508
861
  // object storage and the configured ffmpeg executable are both available.
509
862
  voiceInputResumableEnabled: EnvBoolean.default(true),
510
- voiceInputResumableMaxDurationSeconds: z.coerce.number().int().positive().max(8 * 60 * 60).default(2 * 60 * 60),
511
- voiceInputResumableMaxSizeBytes: z.coerce.number().int().positive().max(512 * 1024 * 1024).default(512 * 1024 * 1024),
512
- voiceInputResumableMaxChunkSizeBytes: z.coerce.number().int().positive().max(25 * 1024 * 1024).default(8 * 1024 * 1024),
513
- voiceInputResumableRetentionSeconds: z.coerce.number().int().positive().max(7 * 24 * 60 * 60).default(24 * 60 * 60),
514
- voiceInputFfmpegPath: z.string().trim().min(1).max(1024).default("ffmpeg"),
863
+ voiceInputResumableMaxDurationSeconds: z2.coerce.number().int().positive().max(8 * 60 * 60).default(2 * 60 * 60),
864
+ voiceInputResumableMaxSizeBytes: z2.coerce.number().int().positive().max(512 * 1024 * 1024).default(512 * 1024 * 1024),
865
+ voiceInputResumableMaxChunkSizeBytes: z2.coerce.number().int().positive().max(25 * 1024 * 1024).default(8 * 1024 * 1024),
866
+ voiceInputResumableRetentionSeconds: z2.coerce.number().int().positive().max(7 * 24 * 60 * 60).default(24 * 60 * 60),
867
+ voiceInputFfmpegPath: z2.string().trim().min(1).max(1024).default("ffmpeg"),
515
868
  // Preferred provider order (comma-separated ids). First configured+ready wins.
516
869
  // Connected subscription STT is preferred by default; operators can put
517
870
  // openai/azure-openai first explicitly.
518
871
  // Supported: supergrok-subscription, codex-subscription, openai, azure-openai.
519
- voiceInputProviderOrder: z.string().default("supergrok-subscription,codex-subscription,openai,azure-openai"),
872
+ voiceInputProviderOrder: z2.string().default("supergrok-subscription,codex-subscription,openai,azure-openai"),
520
873
  // OpenAI public /v1/audio/transcriptions path. Reuses OPENGENI_OPENAI_API_KEY
521
874
  // when voiceInputOpenaiApiKey is unset. Default model is gpt-transcribe.
522
875
  voiceInputOpenaiEnabled: EnvBoolean.default(true),
523
- voiceInputOpenaiApiKey: z.string().optional(),
524
- voiceInputOpenaiBaseUrl: z.string().optional(),
525
- voiceInputOpenaiModel: z.string().default("gpt-transcribe"),
876
+ voiceInputOpenaiApiKey: z2.string().optional(),
877
+ voiceInputOpenaiBaseUrl: z2.string().optional(),
878
+ voiceInputOpenaiModel: z2.string().default("gpt-transcribe"),
526
879
  // Azure OpenAI deployment-scoped audio transcriptions. Reuses the turn-model
527
880
  // Azure endpoint/key/AD token when voice-specific overrides are unset.
528
881
  voiceInputAzureEnabled: EnvBoolean.default(true),
529
- voiceInputAzureEndpoint: z.string().optional(),
530
- voiceInputAzureDeployment: z.string().optional(),
531
- voiceInputAzureApiVersion: z.string().optional(),
532
- voiceInputAzureApiKey: z.string().optional(),
533
- voiceInputAzureAdToken: z.string().optional(),
882
+ voiceInputAzureEndpoint: z2.string().optional(),
883
+ voiceInputAzureDeployment: z2.string().optional(),
884
+ voiceInputAzureApiVersion: z2.string().optional(),
885
+ voiceInputAzureApiKey: z2.string().optional(),
886
+ voiceInputAzureAdToken: z2.string().optional(),
534
887
  // Legacy opt-in for undocumented ChatGPT /backend-api/transcribe. When
535
888
  // OPENGENI_CODEX_SUBSCRIPTION_ENABLED is true, Codex STT is included without
536
889
  // this flag. Set false and omit codex-subscription from PROVIDER_ORDER to
537
890
  // keep subscription model routing while disabling Codex voice input.
538
891
  voiceInputCodexExperimentalEnabled: EnvBoolean.default(false),
539
- modelPricingJson: z.string().default("{}"),
892
+ modelPricingJson: z2.string().default("{}"),
540
893
  // Supported-model membership source. Database mode is resolved by the async
541
894
  // core overlay; getSettings remains synchronous and env-only.
542
- modelCatalogSource: z.enum(["code", "database"]).default("code"),
895
+ modelCatalogSource: z2.enum(["code", "database"]).default("code"),
543
896
  // Deployment-owned workspace-facing price policy. This is deliberately
544
897
  // separate from catalog membership and upstream credential ownership.
545
898
  // Shape: { "product/model-id": "free" | "credits" }.
546
- modelCostPolicyJson: z.string().default("{}"),
899
+ modelCostPolicyJson: z2.string().default("{}"),
547
900
  // Optional per-product agent guidance. Database mode replaces this with the
548
901
  // singleton document's validated modelNotes map.
549
- modelNotesJson: z.string().default("{}"),
902
+ modelNotesJson: z2.string().default("{}"),
550
903
  // Managed OpenRouter credential. The curated model table is injected in
551
904
  // code/catalog-document resolution and never read from host provider JSON.
552
- openrouterApiKey: z.string().optional(),
905
+ openrouterApiKey: z2.string().optional(),
553
906
  // Internal, secret-free catalog overlays populated only by
554
907
  // applyModelCatalogDocument. They intentionally have no OPENGENI_* env
555
908
  // binding so database mode cannot be bypassed with a second source.
556
- resolvedGatewayModelsJson: z.string().optional(),
557
- resolvedOpenRouterModelsJson: z.string().optional(),
909
+ resolvedGatewayModelsJson: z2.string().optional(),
910
+ resolvedOpenRouterModelsJson: z2.string().optional(),
911
+ resolvedCodexModelsJson: z2.string().optional(),
558
912
  // Extra (non-built-in) model providers, declared by the host as a JSON
559
913
  // provider registry. Each entry carries its own base URL, API key, wire API
560
914
  // ("responses" | "chat") and the models it exposes. The models a client may
561
915
  // use are the UNION of the built-in provider's allowed models and every
562
916
  // registry provider's models. validateSettings parses this at boot so a
563
917
  // malformed registry / unresolvable key / id collision fails fast.
564
- modelProvidersJson: z.string().default("[]"),
918
+ modelProvidersJson: z2.string().default("[]"),
565
919
  // Codex (ChatGPT) subscription: when enabled, a per-workspace connected
566
920
  // subscription is injected as a synthetic "codex-subscription" registry
567
921
  // provider whose models route through the ChatGPT backend (@opengeni/codex).
@@ -571,15 +925,17 @@ var SettingsSchema = z.object({
571
925
  // account pool and a distinct rail from the existing xai/* API-key provider.
572
926
  supergrokSubscriptionEnabled: EnvBoolean.default(false),
573
927
  // OPENGENI_SUPERGROK_SUBSCRIPTION_ENABLED
928
+ claudeSubscriptionEnabled: EnvBoolean.default(false),
929
+ // OPENGENI_CLAUDE_SUBSCRIPTION_ENABLED
574
930
  // Maximum silence between complete, valid SuperGrok SSE data events. This is
575
931
  // not a request/run duration cap; every valid event resets the timer.
576
- supergrokResponseStreamIdleTimeoutMs: z.coerce.number().int().positive().max(24 * 60 * 6e4).default(XAI_RESPONSE_STREAM_IDLE_TIMEOUT_MS),
932
+ supergrokResponseStreamIdleTimeoutMs: z2.coerce.number().int().positive().max(24 * 60 * 6e4).default(XAI_RESPONSE_STREAM_IDLE_TIMEOUT_MS),
577
933
  // Expose the connected apps attached to a Codex subscription through the
578
934
  // synthetic codex_apps MCP server. Independent from subscription routing so
579
935
  // operators can use Codex models without exposing ChatGPT connectors.
580
936
  codexConnectedAppsEnabled: EnvBoolean.default(false),
581
937
  // OPENGENI_CODEX_CONNECTED_APPS_ENABLED
582
- codexProductSku: z.string().optional(),
938
+ codexProductSku: z2.string().optional(),
583
939
  // OPENGENI_CODEX_PRODUCT_SKU (X-OpenAI-Product-Sku, apps only)
584
940
  // Progressive MCP disclosure (Codex-CLI-style tool_search): on a codex turn,
585
941
  // flag non-eager selected MCP tools `defer_loading:true` (dropping their
@@ -603,10 +959,23 @@ var SettingsSchema = z.object({
603
959
  // Multi-account P3 (auto-rotation): an account is "near exhaustion" — ineligible to be
604
960
  // rotated TO — when EITHER usage window (5h/weekly) is at/over this percent. Default 90 to
605
961
  // match the UI danger flip (UsageBar danger at pct >= 90). OPENGENI_CODEX_ROTATION_NEAR_EXHAUSTION_PCT.
606
- codexRotationNearExhaustionPct: z.coerce.number().int().min(1).max(100).default(90),
962
+ codexRotationNearExhaustionPct: z2.coerce.number().int().min(1).max(100).default(90),
963
+ reasoningConfigurationUpdatesEnabled: EnvBoolean.default(false),
607
964
  openaiReasoningEffort: ReasoningEffort.default("low"),
608
- openaiAllowedReasoningEfforts: z.string().default("low,medium,high,xhigh,max"),
609
- openaiResponsesTransport: z.enum(["http", "websocket"]).default("http"),
965
+ openaiAllowedReasoningEfforts: z2.string().default("low,medium,high,xhigh,max"),
966
+ // Default for new chats and scheduled tasks when the workspace has an
967
+ // OpenGeni credit balance, no saved workspace default, and no usable
968
+ // connected subscription, and the deployment default is not already a
969
+ // credits-billed model. Selected only when this model is selectable in the
970
+ // workspace catalog; otherwise the first selectable credits model is used.
971
+ // OPENGENI_CREDITS_DEFAULT_MODEL / OPENGENI_CREDITS_DEFAULT_REASONING_EFFORT.
972
+ creditsDefaultModel: z2.string().trim().min(1).max(256).refine((value) => !/[\u000A\u000D|]/u.test(value), {
973
+ message: "credits default model must not contain newlines or the | field separator"
974
+ }).default("gpt-6-luna"),
975
+ // Preferred effort for the credits default. Clamped to the highest effort
976
+ // the selected model supports at or below this value.
977
+ creditsDefaultReasoningEffort: ReasoningEffort.default("xhigh"),
978
+ openaiResponsesTransport: z2.enum(["http", "websocket"]).default("http"),
610
979
  // Provider-assigned item ids (rs_/msg_/fc_…) in Responses API input are
611
980
  // resolved against the provider's server-side response store. That store is
612
981
  // not durable enough to anchor long runs on: a response that streamed fine
@@ -615,7 +984,7 @@ var SettingsSchema = z.object({
615
984
  // every model-call input so requests are self-contained — conversation
616
985
  // truth already lives client-side in session_history_items. "preserve"
617
986
  // keeps the SDK's pass-through behavior.
618
- openaiProviderItemIds: z.enum(["strip", "preserve"]).default("strip"),
987
+ openaiProviderItemIds: z2.enum(["strip", "preserve"]).default("strip"),
619
988
  // With ids stripped the provider cannot resolve prior reasoning server-side,
620
989
  // so request reasoning.encrypted_content and send it back with each call:
621
990
  // reasoning continuity without depending on provider-side storage.
@@ -623,7 +992,7 @@ var SettingsSchema = z.object({
623
992
  // Model-call retry budget for transient provider failures (429s and friends).
624
993
  // The openai client default of 2 retries is too small for sustained TPM
625
994
  // backpressure during long autonomous runs.
626
- openaiMaxRetries: z.coerce.number().int().nonnegative().default(5),
995
+ openaiMaxRetries: z2.coerce.number().int().nonnegative().default(5),
627
996
  // Native hosted web search. The live Azure Responses path executes the
628
997
  // hosted web_search tool, so this is provider-unconditional: ON by default
629
998
  // on every provider, exposed only so operators can disable it. When true,
@@ -631,6 +1000,19 @@ var SettingsSchema = z.object({
631
1000
  // merged with the MCP-server tools (getAllTools = [...mcpTools, ...tools])
632
1001
  // and the sandbox capability tools, never replacing them.
633
1002
  webSearchEnabled: EnvBoolean.default(true),
1003
+ // Jev (TypeSafe's fast judge model) for worker-side agent tools. Without a
1004
+ // usable key every Jev-backed feature is off. The key stays on the server
1005
+ // (API and worker) and never reaches a sandbox or Connected Machine.
1006
+ jevApiKey: z2.string().optional(),
1007
+ jevBaseUrl: z2.string().url().default("https://api.typesafe.ai"),
1008
+ jevModel: z2.string().trim().min(1).max(128).default("jev-latest"),
1009
+ jevRequestTimeoutMs: z2.coerce.number().int().positive().max(12e4).default(1e4),
1010
+ // Jev-backed `code_search` agent tool. `off` never offers it, `opt_in` offers
1011
+ // it only where workspace settings enable it, `default_on` offers it
1012
+ // everywhere except workspaces that disable it, and `experiment` gives it to
1013
+ // a fixed half of sessions in workspaces without their own setting, for
1014
+ // comparison. It also needs jevApiKey.
1015
+ codeSearchMode: z2.enum(["off", "opt_in", "default_on", "experiment"]).default("off"),
634
1016
  // Deployment-default agent persona template (the white-label surface). The
635
1017
  // runtime resolves the effective template per turn as
636
1018
  // per-session-override > per-workspace override > this default, substitutes
@@ -638,38 +1020,40 @@ var SettingsSchema = z.object({
638
1020
  // it when the template omits the marker), and uses the result as the agent's
639
1021
  // instructions. Defaulting to DEFAULT_AGENT_INSTRUCTIONS keeps the composed
640
1022
  // default pinned by runtime tests.
641
- agentInstructionsTemplate: z.string().default(DEFAULT_AGENT_INSTRUCTIONS),
642
- azureOpenaiBaseUrl: z.string().optional(),
643
- azureOpenaiEndpoint: z.string().optional(),
644
- azureOpenaiDeployment: z.string().optional(),
645
- azureOpenaiApiVersion: z.string().optional(),
646
- azureOpenaiApiKey: z.string().optional(),
647
- azureOpenaiAdToken: z.string().optional(),
1023
+ agentInstructionsTemplate: z2.string().default(DEFAULT_AGENT_INSTRUCTIONS),
1024
+ azureOpenaiBaseUrl: z2.string().optional(),
1025
+ azureOpenaiEndpoint: z2.string().optional(),
1026
+ azureOpenaiDeployment: z2.string().optional(),
1027
+ azureOpenaiApiVersion: z2.string().optional(),
1028
+ azureOpenaiApiKey: z2.string().optional(),
1029
+ azureOpenaiAdToken: z2.string().optional(),
648
1030
  disableOpenaiTracing: EnvBoolean.default(false),
649
1031
  sandboxBackend: SandboxBackend.default("docker"),
650
- dockerImage: z.string().default("opengeni-sandbox:local"),
1032
+ dockerImage: z2.string().default("opengeni-sandbox:local"),
651
1033
  // Explicit deployment contract: the configured base sandbox image contains
652
1034
  // the verified, self-contained native artifact runtime at its fixed image
653
1035
  // paths. Disabled by default so arbitrary/custom provider images never make
654
1036
  // document/spreadsheet/presentation skills appear when their runtime is
655
- // absent. Per-pack/per-rig image overrides fail closed in the worker even
656
1037
  // when this base-image contract is enabled.
657
1038
  sandboxArtifactRuntimeEnabled: EnvBoolean.default(false),
658
- dockerExposedPorts: z.string().default(""),
659
- dockerNetwork: z.string().optional(),
1039
+ dockerExposedPorts: z2.string().default(""),
1040
+ dockerNetwork: z2.string().optional(),
660
1041
  // When the worker itself runs in a container and talks to a host Docker daemon,
661
1042
  // this directory must be bind-mounted at the exact same absolute path on both
662
1043
  // sides. The Agents SDK materializes the workspace here before bind-mounting it
663
1044
  // into the sandbox container.
664
- dockerWorkspaceBaseDir: z.string().min(1).optional(),
665
- modalAppName: z.string().default("opengeni-sandbox"),
666
- modalImageRef: z.string().optional(),
1045
+ dockerWorkspaceBaseDir: z2.string().min(1).optional(),
1046
+ modalAppName: z2.string().default("opengeni-sandbox"),
1047
+ modalImageRef: z2.string().optional(),
1048
+ // Activate only after descriptor-aware readers, DB fences and the exact
1049
+ // native image have passed the supervised-command rollout canary.
1050
+ modalCommandSupervisionEnabled: EnvBoolean.default(false),
667
1051
  // Provider-native immutable Modal image ID for the exact logical
668
1052
  // `modalImageRef`. When set, the runtime uses ModalImageSelector.fromId and
669
1053
  // never asks Modal to parse or import the registry ref. The logical ref is
670
1054
  // still persisted on the sandbox lease for provenance and conflict fencing;
671
1055
  // the Modal session envelope persists the actual image ID.
672
- modalImageId: z.string().min(4).max(128).regex(/^im-[A-Za-z0-9]+$/).optional(),
1056
+ modalImageId: z2.string().min(4).max(128).regex(/^im-[A-Za-z0-9]+$/).optional(),
673
1057
  // Name of a Modal Secret (containing REGISTRY_USERNAME + REGISTRY_PASSWORD) used
674
1058
  // to authenticate the pull of `modalImageRef` from a PRIVATE registry. When UNSET
675
1059
  // (the default), the sandbox image is pulled UNAUTHENTICATED — i.e. it must be a
@@ -678,7 +1062,7 @@ var SettingsSchema = z.object({
678
1062
  // run a private image (e.g. a cloud-hosted ACR/ECR/GCR digest): the runtime resolves
679
1063
  // the named Secret and builds the image via `fromRegistry(tag, secret)` before the
680
1064
  // first sandbox is created. Knob: OPENGENI_MODAL_IMAGE_REGISTRY_SECRET.
681
- modalImageRegistrySecret: z.string().optional(),
1065
+ modalImageRegistrySecret: z2.string().optional(),
682
1066
  // Modal's hard sandbox lifetime (timeoutMs = this * 1000), counted from box
683
1067
  // creation. A resume-by-id does NOT reset that provider clock. It is the
684
1068
  // BACKSTOP that reclaims a box if the reaper/worker is down, NOT the warm-window
@@ -689,17 +1073,17 @@ var SettingsSchema = z.object({
689
1073
  // deadline-aware snapshot/rematerialization transition. The transition—not a
690
1074
  // larger timeout—is what lets a session outlive one finite provider box.
691
1075
  // Knob: OPENGENI_MODAL_TIMEOUT_SECONDS.
692
- modalTimeoutSeconds: z.coerce.number().int().positive().max(86400).default(86400),
1076
+ modalTimeoutSeconds: z2.coerce.number().int().positive().max(86400).default(86400),
693
1077
  // Optional provider reservations for every new Modal sandbox. CPU is measured
694
1078
  // in physical cores and may be fractional; memory is measured in MiB. Leave
695
1079
  // both unset to preserve Modal's provider defaults. The Agents adapter stores
696
1080
  // the resolved values in its session state so snapshot replacement and exact
697
1081
  // resume cannot silently change the reservation.
698
- modalSandboxCpu: z.coerce.number().positive().optional(),
699
- modalSandboxMemoryMiB: z.coerce.number().int().positive().optional(),
700
- modalTokenId: z.string().optional(),
701
- modalTokenSecret: z.string().optional(),
702
- modalEnvironment: z.string().optional(),
1082
+ modalSandboxCpu: z2.coerce.number().positive().optional(),
1083
+ modalSandboxMemoryMiB: z2.coerce.number().int().positive().optional(),
1084
+ modalTokenId: z2.string().optional(),
1085
+ modalTokenSecret: z2.string().optional(),
1086
+ modalEnvironment: z2.string().optional(),
703
1087
  // modal gap-fill: idleTimeoutMs + workspacePersistence were unmapped (module 03 §4.1).
704
1088
  //
705
1089
  // CRITICAL (sandbox-file-persistence): when this is UNSET the Modal SDK sends
@@ -717,7 +1101,7 @@ var SettingsSchema = z.object({
717
1101
  // explicitly (OPENGENI_MODAL_IDLE_TIMEOUT_SECONDS) only to deliberately idle-reap
718
1102
  // SOONER than the hard lifetime; the boot invariant forbids a value that would
719
1103
  // reap before reaperPeriod + idleGrace elapses.
720
- modalIdleTimeoutSeconds: z.coerce.number().int().positive().optional(),
1104
+ modalIdleTimeoutSeconds: z2.coerce.number().int().positive().optional(),
721
1105
  // /workspace FILE PERSISTENCE across warm/cold cycles. Directory snapshots
722
1106
  // preserve only the durable user workspace, so provider recovery does not
723
1107
  // restore an entire machine image or replace the selected rig/base image.
@@ -726,16 +1110,14 @@ var SettingsSchema = z.object({
726
1110
  // Modal creations only.
727
1111
  // `snapshot_filesystem` remains available for explicit compatibility and
728
1112
  // immutable rig-image materialization. `tar` is the portable fallback.
729
- modalWorkspacePersistence: z.enum(["tar", "snapshot_filesystem", "snapshot_directory"]).default("snapshot_directory"),
1113
+ modalWorkspacePersistence: z2.enum(["tar", "snapshot_filesystem", "snapshot_directory"]).default("snapshot_directory"),
730
1114
  // Shared desktop toggle: this module reads it for the 6080 port-merge; the
731
1115
  // owner module (P4.x) acts on it to launch the display stack.
732
1116
  sandboxDesktopEnabled: EnvBoolean.default(false),
733
- // Human take-control toggle: when ON (default) the negotiated DesktopStream
734
- // cell advertises mode "interactive" — the noVNC viewer can drive mouse+keyboard
735
- // into :0 (x11vnc runs without -viewonly). Turn it OFF for a genuinely read-only
736
- // deployment: the cell reports mode "read-only" and the client disables the
737
- // "Take control" affordance. This gates the HUMAN viewer plane; agent
738
- // interaction is authorized through managed ComputerSession tools.
1117
+ // Human sandbox input policy. Canonical ComputerSession attachments reflect
1118
+ // it and /actions enforces it independently of the client. The legacy
1119
+ // DesktopStream adapter also reports read-only when disabled. Agent tools
1120
+ // retain their separate session-control authority.
739
1121
  sandboxDesktopInteractive: EnvBoolean.default(true),
740
1122
  // REAL PTY terminal toggle (P5.t): gates the ttyd pty-ws plane (7681) the API
741
1123
  // mints over the SAME tunnel as the desktop. Defaults ON — the interactive
@@ -747,13 +1129,13 @@ var SettingsSchema = z.object({
747
1129
  // The desktop framebuffer geometry the pixel plane advertises + launches the
748
1130
  // display stack with (P4.2). v1 has no live RANDR resize; a change is a full
749
1131
  // down→up restart. Defaults match the proven spike geometry (1280x800).
750
- streamResolutionWidth: z.coerce.number().int().positive().default(1280),
751
- streamResolutionHeight: z.coerce.number().int().positive().default(800),
1132
+ streamResolutionWidth: z2.coerce.number().int().positive().default(1280),
1133
+ streamResolutionHeight: z2.coerce.number().int().positive().default(800),
752
1134
  // Recording loop: ffmpeg x11grab of :0 → mp4/webm → @opengeni/storage.
753
1135
  // recordingMaxBytes caps the in-memory finalize buffer (≤ storage single-PUT);
754
1136
  // recordingMaxSeconds is the ffmpeg -t hard ceiling (bounds a multi-day turn).
755
1137
  recordingEnabled: EnvBoolean.default(true),
756
- recordingDefaultCodec: z.enum(["h264-mp4", "vp9-webm"]).default("h264-mp4"),
1138
+ recordingDefaultCodec: z2.enum(["h264-mp4", "vp9-webm"]).default("h264-mp4"),
757
1139
  // Workbench v2 turn-end workspace capture. When on, the turn
758
1140
  // activity probes the box's changed files off the live box at turn end and
759
1141
  // persists a capture revision (blobs in @opengeni/storage) so the workbench
@@ -761,85 +1143,86 @@ var SettingsSchema = z.object({
761
1143
  // fully behind this flag: off ⇒ capture is skipped and reads fall back to the
762
1144
  // live/wake path (status-quo behavior). Default on; explicit per environment.
763
1145
  workspaceCaptureEnabled: EnvBoolean.default(true),
764
- recordingFramerate: z.coerce.number().int().positive().default(15),
765
- recordingMaxSeconds: z.coerce.number().int().positive().default(600),
766
- recordingMaxBytes: z.coerce.number().int().positive().default(268435456),
1146
+ recordingFramerate: z2.coerce.number().int().positive().default(15),
1147
+ recordingMaxSeconds: z2.coerce.number().int().positive().default(600),
1148
+ recordingMaxBytes: z2.coerce.number().int().positive().default(268435456),
767
1149
  // 256 MB
768
1150
  // --- daytona ---
769
- daytonaApiKey: z.string().optional(),
770
- daytonaApiUrl: z.string().url().optional(),
771
- daytonaTarget: z.string().optional(),
772
- daytonaImage: z.string().optional(),
773
- daytonaSnapshotName: z.string().optional(),
774
- daytonaAutoStopInterval: z.coerce.number().int().nonnegative().optional(),
1151
+ daytonaApiKey: z2.string().optional(),
1152
+ daytonaApiUrl: z2.string().url().optional(),
1153
+ daytonaTarget: z2.string().optional(),
1154
+ daytonaImage: z2.string().optional(),
1155
+ daytonaSnapshotName: z2.string().optional(),
1156
+ daytonaAutoStopInterval: z2.coerce.number().int().nonnegative().optional(),
775
1157
  // 0 disables idle-kill
776
- daytonaTimeoutSeconds: z.coerce.number().int().positive().optional(),
777
- daytonaExposedPortUrlTtlSeconds: z.coerce.number().int().positive().optional(),
1158
+ daytonaTimeoutSeconds: z2.coerce.number().int().positive().optional(),
1159
+ daytonaExposedPortUrlTtlSeconds: z2.coerce.number().int().positive().optional(),
778
1160
  // --- runloop ---
779
- runloopApiKey: z.string().optional(),
780
- runloopBaseUrl: z.string().url().optional(),
781
- runloopBlueprintName: z.string().optional(),
782
- runloopBlueprintId: z.string().optional(),
1161
+ runloopApiKey: z2.string().optional(),
1162
+ runloopBaseUrl: z2.string().url().optional(),
1163
+ runloopBlueprintName: z2.string().optional(),
1164
+ runloopBlueprintId: z2.string().optional(),
783
1165
  runloopTunnel: EnvBoolean.default(true),
784
- runloopKeepAliveSeconds: z.coerce.number().int().positive().optional(),
1166
+ runloopKeepAliveSeconds: z2.coerce.number().int().positive().optional(),
785
1167
  // --- e2b (SDK reads E2B_API_KEY from env; mirrored for validation + forwarding) ---
786
- e2bApiKey: z.string().optional(),
787
- e2bTemplate: z.string().optional(),
788
- e2bTimeoutSeconds: z.coerce.number().int().positive().optional(),
789
- e2bTimeoutAction: z.enum(["pause", "kill"]).optional(),
1168
+ e2bApiKey: z2.string().optional(),
1169
+ e2bTemplate: z2.string().optional(),
1170
+ e2bTimeoutSeconds: z2.coerce.number().int().positive().optional(),
1171
+ e2bTimeoutAction: z2.enum(["pause", "kill"]).optional(),
790
1172
  e2bAllowInternetAccess: EnvBoolean.optional(),
791
1173
  e2bAutoResume: EnvBoolean.optional(),
792
- e2bWorkspacePersistence: z.enum(["tar", "snapshot"]).optional(),
1174
+ e2bWorkspacePersistence: z2.enum(["tar", "snapshot"]).optional(),
793
1175
  // --- blaxel ---
794
- blaxelApiKey: z.string().optional(),
795
- blaxelImage: z.string().optional(),
796
- blaxelRegion: z.string().optional(),
1176
+ blaxelApiKey: z2.string().optional(),
1177
+ blaxelImage: z2.string().optional(),
1178
+ blaxelRegion: z2.string().optional(),
797
1179
  blaxelExposedPortPublic: EnvBoolean.optional(),
798
1180
  // public vs bl_preview_token
799
- blaxelExposedPortUrlTtlSeconds: z.coerce.number().int().positive().optional(),
800
- blaxelMemoryMb: z.coerce.number().int().positive().optional(),
801
- blaxelTtl: z.string().optional(),
1181
+ blaxelExposedPortUrlTtlSeconds: z2.coerce.number().int().positive().optional(),
1182
+ blaxelMemoryMb: z2.coerce.number().int().positive().optional(),
1183
+ blaxelTtl: z2.string().optional(),
802
1184
  // --- cloudflare (headless) ---
803
- cloudflareWorkerUrl: z.string().url().optional(),
804
- cloudflareApiKey: z.string().optional(),
1185
+ cloudflareWorkerUrl: z2.string().url().optional(),
1186
+ cloudflareApiKey: z2.string().optional(),
1187
+ experimentalBrowserContextPoolEnabled: EnvBoolean.default(false),
805
1188
  // --- remote browser placements ---
806
1189
  // Provider credentials are injected only into the placement-resident
807
1190
  // browserd launch. They never enter session contracts, journals, or sandboxes.
808
- browserbaseApiKey: z.string().min(1).max(8192).optional(),
809
- kernelApiKey: z.string().min(1).max(8192).optional(),
810
- kernelEndpoint: z.string().url().optional(),
811
- kernelBrowserTimeoutSeconds: z.coerce.number().int().positive().max(86400).default(3600),
1191
+ browserbaseApiKey: z2.string().min(1).max(8192).optional(),
1192
+ kernelApiKey: z2.string().min(1).max(8192).optional(),
1193
+ kernelEndpoint: z2.string().url().optional(),
1194
+ kernelBrowserTimeoutSeconds: z2.coerce.number().int().positive().max(86400).default(3600),
812
1195
  kernelBrowserStealth: EnvBoolean.default(false),
813
1196
  // --- vercel (headless) ---
814
- vercelToken: z.string().optional(),
815
- vercelProjectId: z.string().optional(),
816
- vercelTeamId: z.string().optional(),
817
- vercelRuntime: z.string().optional(),
1197
+ vercelToken: z2.string().optional(),
1198
+ vercelProjectId: z2.string().optional(),
1199
+ vercelTeamId: z2.string().optional(),
1200
+ vercelRuntime: z2.string().optional(),
818
1201
  // --- OpenSandbox (optional Kubernetes-native provisioned sandbox) ---
819
- openSandboxBaseUrl: z.string().url().optional(),
820
- openSandboxApiKey: z.string().min(1).optional(),
1202
+ openSandboxBaseUrl: z2.string().url().optional(),
1203
+ openSandboxApiKey: z2.string().min(1).optional(),
821
1204
  // Release and preview profiles must provide an immutable OCI digest. The
822
1205
  // adapter refuses tag-only references when this backend is active.
823
- openSandboxImage: z.string().min(1).optional(),
1206
+ openSandboxImage: z2.string().min(1).optional(),
824
1207
  // Renewable provider TTL is a leak/backstop clock, not OpenGeni's idle
825
1208
  // policy. The pinned server accepts a one-minute minimum; ordinary
826
1209
  // deployments default to one hour.
827
- openSandboxTtlSeconds: z.coerce.number().int().min(60).max(86400).default(3600),
1210
+ openSandboxTtlSeconds: z2.coerce.number().int().min(60).max(86400).default(3600),
828
1211
  openSandboxUseServerProxy: EnvBoolean.default(true),
829
1212
  // Channel B (browserd / noVNC / ttyd) uses OSEP-0011 signed URI-mode ingress.
830
1213
  // Exec/files stay on the private lifecycle server-proxy regardless of this flag.
831
1214
  openSandboxSignedEndpoints: EnvBoolean.default(false),
832
- openSandboxSignedEndpointTtlSeconds: z.coerce.number().int().min(60).max(3600).default(600),
833
- openSandboxChannelBPublicBaseUrl: z.string().url().optional(),
1215
+ openSandboxSignedEndpointTtlSeconds: z2.coerce.number().int().min(60).max(3600).default(600),
1216
+ openSandboxChannelBPublicBaseUrl: z2.string().url().optional(),
834
1217
  // Emergency hatch only: force JPEG/RFB through the API frame-proxy even when
835
1218
  // signed endpoints are on (M2 subprotocol failure). Unset means OpenSandbox
836
1219
  // uses the frame-proxy unless signed endpoints are on.
837
1220
  openSandboxInteractionFrameProxy: EnvBoolean.optional(),
838
- openSandboxPoolRef: z.string().regex(/^[A-Za-z0-9](?:[A-Za-z0-9._-]*[A-Za-z0-9])?$/).optional(),
1221
+ openSandboxPoolRef: z2.string().regex(/^[A-Za-z0-9](?:[A-Za-z0-9._-]*[A-Za-z0-9])?$/).optional(),
839
1222
  // Optional same-cluster, read-only observability projection. The application
840
1223
  // chart sets this only on the control worker and mounts a dedicated projected
841
1224
  // service-account token; non-Kubernetes and remote-provider deployments omit it.
842
- openSandboxKubernetesInventoryNamespace: z.string().min(1).max(63).regex(/^[a-z0-9](?:[-a-z0-9]*[a-z0-9])?$/).optional(),
1225
+ openSandboxKubernetesInventoryNamespace: z2.string().min(1).max(63).regex(/^[a-z0-9](?:[-a-z0-9]*[a-z0-9])?$/).optional(),
843
1226
  // --- sandbox ownership inversion (P1.2 rollout flag, default OFF) ---
844
1227
  // The keystone flag for the stateless resume-by-id model. When FALSE the
845
1228
  // agent-turn path is BYTE-FOR-BYTE today's build-and-discard behavior (no
@@ -893,13 +1276,13 @@ var SettingsSchema = z.object({
893
1276
  // when ABSENT and sandboxSelfhostedEnabled is on, the poll route reports the
894
1277
  // credential plane disabled (graceful degrade, mirrors streamTokenSecret). NEVER
895
1278
  // logged. Lives in the opengeni-runtime secret (Helm-clobbered configmap avoided).
896
- enrollmentSigningSecret: z.string().optional(),
1279
+ enrollmentSigningSecret: z2.string().optional(),
897
1280
  // Connect-info the EnrollmentCredentials hand the agent: the NATS server URL(s)
898
1281
  // the agent dials for the control plane, and the relay edge base URL for streams.
899
1282
  // The per-workspace NATS Account creds binding is infra-deferred (M4/relay
900
1283
  // milestone) — the poll returns these endpoints + a placeholder creds field.
901
- selfhostedNatsUrl: z.string().optional(),
902
- selfhostedRelayUrl: z.string().optional(),
1284
+ selfhostedNatsUrl: z2.string().optional(),
1285
+ selfhostedRelayUrl: z2.string().optional(),
903
1286
  // The HMAC secret the control plane signs the agent's relay PRODUCER token with
904
1287
  // (the `ogr_` envelope threaded into EnrollmentCredentials.relayToken; M8b/design
905
1288
  // §10.5). The relay verifies the producer token with the SAME secret. Optional:
@@ -907,10 +1290,10 @@ var SettingsSchema = z.object({
907
1290
  // plane is simply unavailable until configured). Falls back to streamTokenSecret /
908
1291
  // delegationSecret (same HMAC family) so a deployment with a stream-token secret
909
1292
  // needs no second one. NEVER logged. Lives in the opengeni-runtime secret.
910
- selfhostedRelayTokenSecret: z.string().optional(),
1293
+ selfhostedRelayTokenSecret: z2.string().optional(),
911
1294
  // The minisign PUBLIC key the agent pins for self-update verification (handed to
912
1295
  // the agent in EnrollmentCredentials; the SECRET key lives only in CI).
913
- agentUpdatePublicKey: z.string().optional(),
1296
+ agentUpdatePublicKey: z2.string().optional(),
914
1297
  // --- NATS auth-callout tenancy boundary (bring-your-own-compute M-AUTH; design
915
1298
  // §10.1 NATS Accounts per workspace + §17 the isolation smoke) -------------
916
1299
  // nats-server is configured with AUTH CALLOUT: an external agent connects
@@ -927,24 +1310,24 @@ var SettingsSchema = z.object({
927
1310
  // The callout account SIGNING SEED (`SA...`). Both the user JWT and the
928
1311
  // authorization-response JWT are signed by this account key; its public key
929
1312
  // (`A...`) is the `auth_callout.issuer` in the server config. NEVER logged.
930
- selfhostedNatsCalloutAccountSeed: z.string().optional(),
1313
+ selfhostedNatsCalloutAccountSeed: z2.string().optional(),
931
1314
  // The TARGET ACCOUNT NAME the minted user is placed into (the server-config-mode
932
1315
  // `auth_callout.account`, e.g. "APP"). The responder writes it as the minted user
933
1316
  // JWT `aud` so nats-server binds the agent to this account — the SAME account the
934
1317
  // privileged control plane connects into, so exact process request/reply
935
1318
  // routes. Optional; resolveNatsCalloutConfig defaults it to "APP".
936
- selfhostedNatsCalloutAccountName: z.string().optional(),
1319
+ selfhostedNatsCalloutAccountName: z2.string().optional(),
937
1320
  // The callout RESPONDER's own NATS login (one of the `auth_callout.auth_users`
938
1321
  // in the AUTH account) — the responder connects with this to subscribe
939
1322
  // $SYS.REQ.USER.AUTH. Username/password.
940
- selfhostedNatsCalloutUser: z.string().optional(),
941
- selfhostedNatsCalloutPassword: z.string().optional(),
1323
+ selfhostedNatsCalloutUser: z2.string().optional(),
1324
+ selfhostedNatsCalloutPassword: z2.string().optional(),
942
1325
  // The PRIVILEGED control-plane login (api/worker): a static account user that may
943
1326
  // request exact process RPC subjects + receive their inbox replies. The event bus + the
944
1327
  // selfhosted control RPC ride THIS connection. Username/password; when unset the
945
1328
  // bus connects anonymously (local dev / a NATS with no auth_callout).
946
- selfhostedNatsControlUser: z.string().optional(),
947
- selfhostedNatsControlPassword: z.string().optional(),
1329
+ selfhostedNatsControlUser: z2.string().optional(),
1330
+ selfhostedNatsControlPassword: z2.string().optional(),
948
1331
  // --- selfhosted (Connected Machine) control/exec op deadlines ---------------
949
1332
  // CONTROL ops (ping / fs / git / desktop / pty) keep a short request timeout so
950
1333
  // liveness failures surface promptly. EXEC duration is a different concern: by
@@ -954,8 +1337,8 @@ var SettingsSchema = z.object({
954
1337
  // process deadline by setting a positive value; 0 never schedules a process kill.
955
1338
  // Knobs: OPENGENI_SANDBOX_SELFHOSTED_EXEC_TIMEOUT_MS (default 0 = none) and
956
1339
  // OPENGENI_SANDBOX_SELFHOSTED_CONTROL_TIMEOUT_MS (default 30s).
957
- sandboxSelfhostedExecTimeoutMs: z.coerce.number().int().nonnegative().default(0),
958
- sandboxSelfhostedControlTimeoutMs: z.coerce.number().int().positive().default(3e4),
1340
+ sandboxSelfhostedExecTimeoutMs: z2.coerce.number().int().nonnegative().default(0),
1341
+ sandboxSelfhostedControlTimeoutMs: z2.coerce.number().int().positive().default(3e4),
959
1342
  // --- sandbox lease cadences (cadence invariant validated at boot below) ---
960
1343
  // reaperPeriod < viewerHolderTTL, and reaperPeriod + idleGrace < the EFFECTIVE
961
1344
  // box idle timeout (effectiveModalIdleTimeoutSeconds, which defaults to the hard
@@ -963,12 +1346,12 @@ var SettingsSchema = z.object({
963
1346
  // idle timeout — which we pin high enough (via the idle-timeout default) that
964
1347
  // OpenGeni's reaper, not Modal's idle-reap, governs teardown so /workspace is
965
1348
  // snapshotted before the box dies (sandbox-file-persistence).
966
- sandboxLeaseReaperPeriodMs: z.coerce.number().int().positive().default(3e4),
967
- sandboxViewerHolderTtlMs: z.coerce.number().int().positive().default(9e4),
1349
+ sandboxLeaseReaperPeriodMs: z2.coerce.number().int().positive().default(3e4),
1350
+ sandboxViewerHolderTtlMs: z2.coerce.number().int().positive().default(9e4),
968
1351
  // A BrowserSession controller refreshes its durable resource and exact
969
1352
  // interaction lease holder together. This longer crash horizon tolerates API
970
1353
  // replacement while still releasing a placement whose controller died.
971
- sandboxInteractionHolderTtlMs: z.coerce.number().int().positive().default(18e4),
1354
+ sandboxInteractionHolderTtlMs: z2.coerce.number().int().positive().default(18e4),
972
1355
  // The DRAIN grace: how long a refcount-0 (draining) lease stays WARM before the
973
1356
  // reaper resume-by-ids the box and terminates it. This is the cost-vs-snappiness
974
1357
  // dial — when the user navigates away the box keeps refcount 0, but it survives
@@ -978,7 +1361,29 @@ var SettingsSchema = z.object({
978
1361
  // fresh EMPTY box; lower it to trade warm cost for a snappier reclaim.
979
1362
  // getSettings caps the default at half a shorter configured Modal lifetime so
980
1363
  // the entire reaper window always fits. Knob: OPENGENI_SANDBOX_IDLE_GRACE_MS.
981
- sandboxIdleGraceMs: z.coerce.number().int().positive().default(9e5),
1364
+ sandboxIdleGraceMs: z2.coerce.number().int().positive().default(9e5),
1365
+ // Idle command containment. A legacy retained background command (a dev
1366
+ // server, a command whose output is still draining, a stopped command the
1367
+ // provider no longer answers for) keeps its box warm through a non-expiring
1368
+ // process holder, so the zero-holder drain never runs and the box would stay
1369
+ // up until the provider deadline kills it uncaptured. Once every session of
1370
+ // the sandbox group has been unused this long - no open turn or pending
1371
+ // request, no held wait_for_input, no other holder or writer, measured from
1372
+ // durable turn/holder/admission facts - the reaper checkpoints the workspace,
1373
+ // stops the box and settles those commands `lost` with reason
1374
+ // `idle_containment`. Default 30min: twice the default
1375
+ // idle grace, so a lease that is only waiting for a "glanced away" user is
1376
+ // never contained earlier than an idle lease would drain, and well inside
1377
+ // the 1h provider-deadline rotation lead, so an idle box is saved long before
1378
+ // the deadline path has to act. Set 0 to disable new idle enrollments without
1379
+ // disabling provider-deadline containment or cancelling an enrolled drain.
1380
+ // A positive window must exceed OPENGENI_SANDBOX_IDLE_GRACE_MS;
1381
+ // an explicit positive value must also stay below OPENGENI_SANDBOX_ROTATION_LEAD_MS,
1382
+ // and with an explicit OPENGENI_MODAL_IDLE_TIMEOUT_SECONDS the reaper period
1383
+ // plus this window plus the drain capture budget must fit before it.
1384
+ // getSettings derives the unset default between those two for short-lived
1385
+ // provider lifetimes. Knob: OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS.
1386
+ sandboxIdleCommandContainmentMs: z2.coerce.number().int().nonnegative().optional().transform((value) => value === 0 ? void 0 : value),
982
1387
  // MID-SESSION /workspace snapshot cadence (sandbox-file-persistence). The
983
1388
  // reaper's drain-persist only protects boxes the reaper itself kills; a box
984
1389
  // that dies any other way (Modal's hard creation-time timeout on a session
@@ -988,20 +1393,20 @@ var SettingsSchema = z.object({
988
1393
  // one (same epoch-fenced fold-onto-lease seam as the drain), bounding the
989
1394
  // worst-case loss of ANY unclean box death to this window. 0 disables.
990
1395
  // Knob: OPENGENI_SANDBOX_SNAPSHOT_INTERVAL_MS. Default 15min.
991
- sandboxSnapshotIntervalMs: z.coerce.number().int().min(0).default(9e5),
1396
+ sandboxSnapshotIntervalMs: z2.coerce.number().int().min(0).default(9e5),
992
1397
  // Maximum time a best-effort /workspace snapshot capture may hold turn/reaper
993
1398
  // cleanup. A hung provider snapshot must never pin a lease holder, block
994
1399
  // graceful shutdown, or become permission to GC an older archive. Timeout is
995
1400
  // treated exactly like a failed best-effort snapshot. Knob:
996
1401
  // OPENGENI_SANDBOX_SNAPSHOT_TIMEOUT_MS. Default 60s.
997
- sandboxSnapshotTimeoutMs: z.coerce.number().int().positive().max(SANDBOX_SNAPSHOT_MAX_TIMEOUT_MS).default(6e4),
1402
+ sandboxSnapshotTimeoutMs: z2.coerce.number().int().positive().max(SANDBOX_SNAPSHOT_MAX_TIMEOUT_MS).default(6e4),
998
1403
  // A zero-holder drain may need substantially longer than a best-effort
999
1404
  // mid-turn/turn-end snapshot for a very large workspace. Keep that provider
1000
1405
  // budget independent so increasing drain recovery headroom cannot pin an
1001
1406
  // ordinary turn finalizer for the same duration. Unset preserves the legacy
1002
1407
  // single-budget behavior. Knob:
1003
1408
  // OPENGENI_SANDBOX_DRAIN_SNAPSHOT_TIMEOUT_MS.
1004
- sandboxDrainSnapshotTimeoutMs: z.coerce.number().int().positive().max(SANDBOX_SNAPSHOT_MAX_TIMEOUT_MS).optional(),
1409
+ sandboxDrainSnapshotTimeoutMs: z2.coerce.number().int().positive().max(SANDBOX_SNAPSHOT_MAX_TIMEOUT_MS).optional(),
1005
1410
  // Begin a controlled snapshot/quiesce/drain/rematerialize transition this far
1006
1411
  // ahead of a finite provider deadline. Modal's 24h creation clock cannot be
1007
1412
  // extended; the logical sandbox outlives it by moving to one successor box.
@@ -1010,22 +1415,22 @@ var SettingsSchema = z.object({
1010
1415
  // an extra coupled environment override. An explicit value may be larger when
1011
1416
  // an operator deliberately wants more rotation headroom; the boot invariant
1012
1417
  // still requires it to remain below the provider lifetime.
1013
- sandboxRotationLeadMs: z.coerce.number().int().positive().default(36e5),
1418
+ sandboxRotationLeadMs: z2.coerce.number().int().positive().default(36e5),
1014
1419
  // Bound provider-deadline rotation admission independently of execution.
1015
1420
  // Every admitted box receives its own durable drain child; the control worker
1016
1421
  // limits provider I/O to 32 concurrent activities. Matching that bound avoids
1017
1422
  // both the old one-box-per-tick deadline backlog and an unbounded provider/API
1018
1423
  // burst. Operators may tune this for a differently-sized worker pool.
1019
- sandboxRotationBatchSize: z.coerce.number().int().positive().max(500).default(32),
1424
+ sandboxRotationBatchSize: z2.coerce.number().int().positive().max(500).default(32),
1020
1425
  // expires_at refresh window for a held lease (>> the turn 10s heartbeat so a
1021
1426
  // single missed heartbeat never TTL-reaps a live turn). The warming TTL is the
1022
1427
  // window a cold->warming spawner has to commit warm before a reaper resets it.
1023
- sandboxLeaseTtlMs: z.coerce.number().int().positive().default(9e4),
1024
- sandboxLeaseWarmingTtlMs: z.coerce.number().int().positive().default(12e4),
1428
+ sandboxLeaseTtlMs: z2.coerce.number().int().positive().default(9e4),
1429
+ sandboxLeaseWarmingTtlMs: z2.coerce.number().int().positive().default(12e4),
1025
1430
  // Overall user-facing budget for warming a sandbox lease. Unlike the lease TTL
1026
1431
  // (a liveness/reaper cadence), this bounds how long one turn waits for capacity
1027
1432
  // or provider creation before surfacing a clear turn.failed error.
1028
- sandboxWarmingTimeoutMs: z.coerce.number().int().positive().default(6e5),
1433
+ sandboxWarmingTimeoutMs: z2.coerce.number().int().positive().default(6e5),
1029
1434
  // Request-scoped workspace control-prefix budget: how long one HTTP-originated
1030
1435
  // session/workspace mutation (Send, Steer, Pause/Resume/Cancel, queue
1031
1436
  // move/edit/delete, composer draft, settings narrowing, quiescent tree
@@ -1034,114 +1439,132 @@ var SettingsSchema = z.object({
1034
1439
  // settlement and claims never use it. The API installs the validated value
1035
1440
  // into @opengeni/db once at app construction; nothing reads the env per
1036
1441
  // request. Env: OPENGENI_WORKSPACE_CONTROL_LOCK_TIMEOUT_MS. Default 20 s.
1037
- workspaceControlLockTimeoutMs: z.coerce.number({
1442
+ workspaceControlLockTimeoutMs: z2.coerce.number({
1038
1443
  message: "OPENGENI_WORKSPACE_CONTROL_LOCK_TIMEOUT_MS must be a positive integer (ms)"
1039
1444
  }).int("OPENGENI_WORKSPACE_CONTROL_LOCK_TIMEOUT_MS must be a positive integer (ms)").positive("OPENGENI_WORKSPACE_CONTROL_LOCK_TIMEOUT_MS must be a positive integer (ms)").default(2e4),
1040
1445
  // Rig setup-script budget (M3): the wall-clock timeout the rig-setup lifecycle
1041
1446
  // hook runs its script under, distinct from the 120s per-command lifecycle
1042
1447
  // default (a rig may compile/install heavy tooling on first cold create).
1043
1448
  // Env: OPENGENI_RIG_SETUP_TIMEOUT_MS. Default 10min.
1044
- rigSetupTimeoutMs: z.coerce.number().int().positive().default(6e5),
1449
+ rigSetupTimeoutMs: z2.coerce.number().int().positive().default(6e5),
1045
1450
  // --- sandbox warm-time billing (P2.1) ---
1046
1451
  // Per-backend warm rate (usd_micros/sec), like modelPricingJson: an empty {}
1047
1452
  // means warm-cost is not debited (warm-seconds are still metered for audit).
1048
1453
  // Shape: { "modal": 5, "runloop": 4, ... }. Backends absent here meter
1049
1454
  // warm-seconds but accrue NO warm_cost / debit (rate 0).
1050
- sandboxWarmRateMicrosPerSecondJson: z.string().default("{}"),
1455
+ sandboxWarmRateMicrosPerSecondJson: z2.string().default("{}"),
1456
+ sandboxWarmBillingMode: z2.enum(["usage_only", "shadow", "credits"]).default("usage_only"),
1051
1457
  // Per-workspace warm cap (cumulative warm-seconds since the start of the UTC
1052
1458
  // month, summed over sandbox.warm_seconds). 0 = unbounded. A workspace over the
1053
1459
  // cap force-drains its VIEWER-ONLY boxes (guarded AND turn_holders=0 — a paying
1054
1460
  // turn is never killed); the reaper then stop()s at refcount 0.
1055
- sandboxMaxWarmSecondsPerWorkspace: z.coerce.number().int().nonnegative().default(0),
1056
- sandboxPreparationProfiles: z.string().default("none"),
1057
- sandboxEnvAllowlist: z.string().default(""),
1058
- objectStorageEndpoint: z.string().url().optional(),
1059
- objectStorageInternalEndpoint: z.string().url().optional(),
1060
- objectStorageSandboxEndpoint: z.string().url().optional(),
1061
- objectStorageBackend: z.enum(["s3-compatible", "aws-s3", "azure-blob", "gcs"]).default("s3-compatible"),
1062
- objectStorageBucket: z.string().min(1).default("opengeni-files"),
1063
- objectStorageRegion: z.string().min(1).default("us-east-1"),
1064
- objectStorageS3Provider: z.string().min(1).default("Minio"),
1065
- objectStorageAccessKeyId: z.string().optional(),
1066
- objectStorageSecretAccessKey: z.string().optional(),
1461
+ sandboxMaxWarmSecondsPerWorkspace: z2.coerce.number().int().nonnegative().default(0),
1462
+ sandboxPreparationProfiles: z2.string().default("none"),
1463
+ sandboxEnvAllowlist: z2.string().default(""),
1464
+ // Comma-separated image references a workspace may select as its default
1465
+ // sandbox image. Empty keeps every workspace on the deployment image.
1466
+ sandboxImageAllowlist: z2.string().default(""),
1467
+ objectStorageEndpoint: z2.string().url().optional(),
1468
+ objectStorageInternalEndpoint: z2.string().url().optional(),
1469
+ objectStorageSandboxEndpoint: z2.string().url().optional(),
1470
+ objectStorageBackend: z2.enum(["s3-compatible", "aws-s3", "azure-blob", "gcs"]).default("s3-compatible"),
1471
+ objectStorageBucket: z2.string().min(1).default("opengeni-files"),
1472
+ objectStorageRegion: z2.string().min(1).default("us-east-1"),
1473
+ objectStorageS3Provider: z2.string().min(1).default("Minio"),
1474
+ objectStorageAccessKeyId: z2.string().optional(),
1475
+ objectStorageSecretAccessKey: z2.string().optional(),
1067
1476
  objectStorageForcePathStyle: EnvBoolean.default(true),
1068
- objectStorageAzureConnectionString: z.string().optional(),
1069
- objectStorageAzureAccountName: z.string().optional(),
1070
- objectStorageAzureAccountKey: z.string().optional(),
1071
- objectStorageAzureEndpoint: z.string().url().optional(),
1072
- objectStorageGcsProjectId: z.string().optional(),
1073
- objectStorageGcsCredentialsJson: z.string().optional(),
1074
- objectStorageGcsKeyFilename: z.string().optional(),
1075
- objectStorageGcsApiEndpoint: z.string().url().optional(),
1076
- documentParser: z.string().min(1).default("liteparse"),
1077
- documentChunkSize: z.coerce.number().int().positive().default(1200),
1078
- documentChunkOverlap: z.coerce.number().int().nonnegative().default(160),
1079
- documentEmbeddingProvider: z.enum(["openai", "deterministic"]).default("openai"),
1080
- documentEmbeddingModel: z.string().min(1).default("text-embedding-3-large"),
1081
- documentEmbeddingDimensions: z.coerce.number().int().positive().default(3072),
1082
- documentEmbeddingApiKey: z.string().optional(),
1083
- documentEmbeddingBaseUrl: z.string().url().optional(),
1084
- documentCurationProvider: z.enum(["openai", "heuristic", "none"]).default("openai"),
1085
- documentCurationModel: z.string().min(1).default("gpt-4o-mini"),
1086
- documentCurationApiKey: z.string().optional(),
1087
- documentCurationBaseUrl: z.string().url().optional(),
1088
- gitAuthorName: z.string().optional(),
1089
- gitAuthorEmail: z.string().optional(),
1090
- gitCommitterName: z.string().optional(),
1091
- gitCommitterEmail: z.string().optional(),
1092
- githubAppManifestBaseUrl: z.string().optional(),
1093
- githubAppManifestStateSecret: z.string().optional(),
1094
- githubAppId: z.string().optional(),
1095
- githubClientId: z.string().optional(),
1096
- githubClientSecret: z.string().optional(),
1477
+ objectStorageAzureConnectionString: z2.string().optional(),
1478
+ objectStorageAzureAccountName: z2.string().optional(),
1479
+ objectStorageAzureAccountKey: z2.string().optional(),
1480
+ objectStorageAzureEndpoint: z2.string().url().optional(),
1481
+ objectStorageGcsProjectId: z2.string().optional(),
1482
+ objectStorageGcsCredentialsJson: z2.string().optional(),
1483
+ objectStorageGcsKeyFilename: z2.string().optional(),
1484
+ objectStorageGcsApiEndpoint: z2.string().url().optional(),
1485
+ documentParser: z2.string().min(1).default("liteparse"),
1486
+ documentChunkSize: z2.coerce.number().int().positive().default(1200),
1487
+ documentChunkOverlap: z2.coerce.number().int().nonnegative().default(160),
1488
+ documentEmbeddingProvider: z2.enum(["openai", "deterministic"]).default("openai"),
1489
+ documentEmbeddingModel: z2.string().min(1).default("text-embedding-3-large"),
1490
+ documentEmbeddingDimensions: z2.coerce.number().int().positive().default(3072),
1491
+ documentEmbeddingApiKey: z2.string().optional(),
1492
+ documentEmbeddingBaseUrl: z2.string().url().optional(),
1493
+ documentEmbeddingBillingMode: z2.enum(["usage_only", "shadow", "credits"]).default("usage_only"),
1494
+ // Explicit commercial cutover. Existing queued revisions remain unpriced;
1495
+ // only jobs created at or after this instant may enter the paid rail.
1496
+ documentEmbeddingCreditsActivatedAt: z2.iso.datetime().optional(),
1497
+ // Customer tariff for the exact UTF-8 bytes sent to the embedder. Zero leaves
1498
+ // the paid mode unavailable; no commercial rate is selected by this PR.
1499
+ documentEmbeddingRateMicrosPerMillionBytes: z2.coerce.number().int().nonnegative().default(0),
1500
+ documentCurationProvider: z2.enum(["openai", "heuristic", "none"]).default("openai"),
1501
+ documentCurationModel: z2.string().min(1).default("gpt-4o-mini"),
1502
+ documentCurationApiKey: z2.string().optional(),
1503
+ documentCurationBaseUrl: z2.string().url().optional(),
1504
+ gitAuthorName: z2.string().optional(),
1505
+ gitAuthorEmail: z2.string().optional(),
1506
+ gitCommitterName: z2.string().optional(),
1507
+ gitCommitterEmail: z2.string().optional(),
1508
+ githubAppManifestBaseUrl: z2.string().optional(),
1509
+ githubAppManifestStateSecret: z2.string().optional(),
1510
+ githubAppId: z2.string().optional(),
1511
+ githubClientId: z2.string().optional(),
1512
+ githubClientSecret: z2.string().optional(),
1097
1513
  /** Default-off rollout for the in-process GitHub repository API tool surface. */
1098
1514
  githubRestMcpEnabled: EnvBoolean.default(false),
1099
1515
  githubPersonalOauthEnabled: EnvBoolean.default(false),
1100
- githubPersonalOauthClientId: z.string().optional(),
1101
- githubPersonalOauthClientSecret: z.string().optional(),
1102
- githubAppSlug: z.string().optional(),
1103
- githubWebhookSecret: z.string().optional(),
1104
- githubAppPrivateKey: z.string().optional(),
1105
- prReviewGithubAppId: z.string().optional(),
1106
- prReviewGithubClientId: z.string().optional(),
1107
- prReviewGithubClientSecret: z.string().optional(),
1108
- prReviewGithubAppSlug: z.string().optional(),
1109
- prReviewGithubWebhookSecret: z.string().optional(),
1110
- prReviewGithubAppPrivateKey: z.string().optional(),
1111
- betterAuthSecret: z.string().optional(),
1112
- betterAuthAllowedHosts: z.string().default(""),
1113
- betterAuthCookieDomain: z.string().optional(),
1114
- betterAuthTrustedOrigins: z.string().default(""),
1115
- managedAuthGoogleClientId: z.string().optional(),
1116
- managedAuthGoogleClientSecret: z.string().optional(),
1117
- managedAuthGithubClientId: z.string().optional(),
1118
- managedAuthGithubClientSecret: z.string().optional(),
1516
+ githubPersonalOauthClientId: z2.string().optional(),
1517
+ githubPersonalOauthClientSecret: z2.string().optional(),
1518
+ githubAppSlug: z2.string().optional(),
1519
+ githubWebhookSecret: z2.string().optional(),
1520
+ githubAppPrivateKey: z2.string().optional(),
1521
+ prReviewGithubAppId: z2.string().optional(),
1522
+ prReviewGithubClientId: z2.string().optional(),
1523
+ prReviewGithubClientSecret: z2.string().optional(),
1524
+ prReviewGithubAppSlug: z2.string().optional(),
1525
+ prReviewGithubWebhookSecret: z2.string().optional(),
1526
+ prReviewGithubAppPrivateKey: z2.string().optional(),
1527
+ betterAuthSecret: z2.string().optional(),
1528
+ betterAuthAllowedHosts: z2.string().default(""),
1529
+ betterAuthCookieDomain: z2.string().optional(),
1530
+ betterAuthTrustedOrigins: z2.string().default(""),
1531
+ managedAuthGoogleClientId: z2.string().optional(),
1532
+ managedAuthGoogleClientSecret: z2.string().optional(),
1533
+ managedAuthGithubClientId: z2.string().optional(),
1534
+ managedAuthGithubClientSecret: z2.string().optional(),
1119
1535
  // Rolling browser login-slot compatibility. Repository/deployment default is
1120
1536
  // deliberately legacy; changing to broker is an operator-authorized rollout.
1121
- managedAuthSessionSetMode: z.enum(["legacy", "dual", "broker"]).default("legacy"),
1537
+ allowedUserEmails: z2.array(z2.email().transform((email) => email.toLowerCase())).min(1).optional(),
1538
+ managedAuthSessionSetMode: z2.enum(["legacy", "dual", "broker"]).default("legacy"),
1539
+ // Deployment ceiling for new managed accounts. When false, managed auth
1540
+ // refuses every new Better Auth account (email/password sign-up and implicit
1541
+ // Google/GitHub sign-up) while existing sign-in, sessions, password reset,
1542
+ // email verification, and invitation-bound account setup keep working. Read
1543
+ // at startup; the 0585 runtime switch pauses sign-ups without a restart.
1544
+ managedAuthNewSignupsEnabled: EnvBoolean.default(true),
1122
1545
  // Query transport is an explicit second-stage rollout. A pre-compatibility
1123
1546
  // web image understands only fragment bearers, so API replicas must keep
1124
1547
  // generating fragment links until the compatible web fleet has converged.
1125
- organizationUserSetupEmailTokenTransport: z.enum(["fragment", "query"]).default("fragment"),
1548
+ organizationUserSetupEmailTokenTransport: z2.enum(["fragment", "query"]).default("fragment"),
1126
1549
  // Query-bearing setup links may appear in controller/edge error logs even
1127
1550
  // when access logs and request tracing are disabled. This explicit operator
1128
1551
  // confirmation keeps query transport fail closed until that separate sink is
1129
1552
  // proven sanitized.
1130
1553
  organizationUserSetupQueryEdgeSanitizationConfirmed: EnvBoolean.default(false),
1131
- resendApiKey: z.string().optional(),
1132
- emailFrom: z.string().default("OpenGeni <auth@mail.opengeni.ai>"),
1133
- stripeSecretKey: z.string().optional(),
1134
- stripePublishableKey: z.string().optional(),
1135
- stripeWebhookSecret: z.string().optional(),
1136
- stripeCreditsProductId: z.string().optional(),
1137
- mcpServers: z.array(
1138
- z.object({
1139
- id: z.string().min(1).regex(registryId),
1140
- name: z.string().min(1).optional(),
1141
- url: z.string().url(),
1142
- allowedTools: z.array(z.string().min(1)).optional(),
1143
- timeoutMs: z.number().int().positive().optional(),
1144
- cacheToolsList: z.boolean().default(false),
1554
+ resendApiKey: z2.string().optional(),
1555
+ emailFrom: z2.string().default("OpenGeni <auth@mail.opengeni.ai>"),
1556
+ stripeSecretKey: z2.string().optional(),
1557
+ stripePublishableKey: z2.string().optional(),
1558
+ stripeWebhookSecret: z2.string().optional(),
1559
+ stripeCreditsProductId: z2.string().optional(),
1560
+ mcpServers: z2.array(
1561
+ z2.object({
1562
+ id: z2.string().min(1).regex(registryId),
1563
+ name: z2.string().min(1).optional(),
1564
+ url: z2.string().url(),
1565
+ allowedTools: z2.array(z2.string().min(1)).optional(),
1566
+ timeoutMs: z2.number().int().positive().optional(),
1567
+ cacheToolsList: z2.boolean().default(false),
1145
1568
  /** Runtime approval policy, overlaid from an attempt-frozen session snapshot. */
1146
1569
  requireApproval: SessionMcpApprovalPolicy.optional(),
1147
1570
  /**
@@ -1150,11 +1573,109 @@ var SettingsSchema = z.object({
1150
1573
  * encrypted capability-installation credentials; do not put secrets in
1151
1574
  * OPENGENI_MCP_SERVERS.
1152
1575
  */
1153
- headers: z.record(z.string(), z.string()).optional(),
1154
- connectionRef: McpServerConnectionRefSchema.optional()
1576
+ headers: z2.record(z2.string(), z2.string()).optional(),
1577
+ connectionRef: McpServerConnectionRefSchema.optional(),
1578
+ operationRecovery: McpOperationRecoverySchema.optional()
1155
1579
  })
1156
1580
  ).default([])
1157
1581
  });
1582
+ var ArtifactServiceSettingsSchema = SettingsSchema.pick({
1583
+ serviceName: true,
1584
+ environment: true,
1585
+ deploymentRevision: true,
1586
+ dbSchema: true,
1587
+ rlsStrategy: true,
1588
+ observabilityStructuredLogs: true,
1589
+ observabilityMetricsEnabled: true,
1590
+ observabilityOtlpEndpoint: true,
1591
+ observabilityOtlpHeaders: true
1592
+ });
1593
+ var ArtifactOutboxSettingsSchema = ArtifactServiceSettingsSchema.extend({
1594
+ natsUrl: SettingsSchema.shape.natsUrl,
1595
+ selfhostedNatsControlUser: SettingsSchema.shape.selfhostedNatsControlUser,
1596
+ selfhostedNatsControlPassword: SettingsSchema.shape.selfhostedNatsControlPassword
1597
+ });
1598
+ var ObjectStorageSettingsSchema = SettingsSchema.pick({
1599
+ objectStorageBackend: true,
1600
+ objectStorageEndpoint: true,
1601
+ objectStorageInternalEndpoint: true,
1602
+ objectStorageSandboxEndpoint: true,
1603
+ objectStorageBucket: true,
1604
+ objectStorageRegion: true,
1605
+ objectStorageAccessKeyId: true,
1606
+ objectStorageSecretAccessKey: true,
1607
+ objectStorageForcePathStyle: true,
1608
+ objectStorageAzureConnectionString: true,
1609
+ objectStorageAzureAccountName: true,
1610
+ objectStorageAzureAccountKey: true,
1611
+ objectStorageAzureEndpoint: true,
1612
+ objectStorageGcsProjectId: true,
1613
+ objectStorageGcsCredentialsJson: true,
1614
+ objectStorageGcsKeyFilename: true,
1615
+ objectStorageGcsApiEndpoint: true
1616
+ });
1617
+ var ArtifactMaterializerSettingsSchema = ArtifactServiceSettingsSchema.extend(
1618
+ ObjectStorageSettingsSchema.shape
1619
+ );
1620
+ function artifactServiceEnvironment(source) {
1621
+ const optional = (name) => optionalEnvironmentValue(name, source);
1622
+ return {
1623
+ serviceName: optional("OPENGENI_SERVICE_NAME"),
1624
+ environment: optional("OPENGENI_ENVIRONMENT"),
1625
+ deploymentRevision: optional("OPENGENI_DEPLOYMENT_REVISION") ?? optional("SOURCE_VERSION") ?? optional("GITHUB_SHA"),
1626
+ dbSchema: optional("OPENGENI_DB_SCHEMA"),
1627
+ rlsStrategy: optional("OPENGENI_RLS_STRATEGY"),
1628
+ observabilityStructuredLogs: optional("OPENGENI_OBSERVABILITY_STRUCTURED_LOGS"),
1629
+ observabilityMetricsEnabled: optional("OPENGENI_OBSERVABILITY_METRICS_ENABLED"),
1630
+ observabilityOtlpEndpoint: optional("OPENGENI_OTEL_EXPORTER_OTLP_ENDPOINT") ?? optional("OTEL_EXPORTER_OTLP_ENDPOINT"),
1631
+ observabilityOtlpHeaders: optional("OPENGENI_OTEL_EXPORTER_OTLP_HEADERS") ?? optional("OTEL_EXPORTER_OTLP_HEADERS")
1632
+ };
1633
+ }
1634
+ function getArtifactOutboxSettings(source = process.env) {
1635
+ const settings = ArtifactOutboxSettingsSchema.parse({
1636
+ ...artifactServiceEnvironment(source),
1637
+ natsUrl: optionalEnvironmentValue("OPENGENI_NATS_URL", source),
1638
+ selfhostedNatsControlUser: optionalEnvironmentValue(
1639
+ "OPENGENI_SELFHOSTED_NATS_CONTROL_USER",
1640
+ source
1641
+ ),
1642
+ selfhostedNatsControlPassword: optionalEnvironmentValue(
1643
+ "OPENGENI_SELFHOSTED_NATS_CONTROL_PASSWORD",
1644
+ source
1645
+ )
1646
+ });
1647
+ dbSearchPath(settings);
1648
+ if (Boolean(settings.selfhostedNatsControlUser) !== Boolean(settings.selfhostedNatsControlPassword)) {
1649
+ throw new Error("Artifact outbox NATS control user and password must be configured together");
1650
+ }
1651
+ return settings;
1652
+ }
1653
+ function getArtifactMaterializerSettings(source = process.env) {
1654
+ const optional = (name) => optionalEnvironmentValue(name, source);
1655
+ const settings = ArtifactMaterializerSettingsSchema.parse({
1656
+ ...artifactServiceEnvironment(source),
1657
+ objectStorageEndpoint: optional("OPENGENI_OBJECT_STORAGE_ENDPOINT"),
1658
+ objectStorageInternalEndpoint: optional("OPENGENI_OBJECT_STORAGE_INTERNAL_ENDPOINT"),
1659
+ objectStorageSandboxEndpoint: optional("OPENGENI_OBJECT_STORAGE_SANDBOX_ENDPOINT"),
1660
+ objectStorageBackend: optional("OPENGENI_OBJECT_STORAGE_BACKEND"),
1661
+ objectStorageBucket: optional("OPENGENI_OBJECT_STORAGE_BUCKET"),
1662
+ objectStorageRegion: optional("OPENGENI_OBJECT_STORAGE_REGION"),
1663
+ objectStorageAccessKeyId: optional("OPENGENI_OBJECT_STORAGE_ACCESS_KEY_ID"),
1664
+ objectStorageSecretAccessKey: optional("OPENGENI_OBJECT_STORAGE_SECRET_ACCESS_KEY"),
1665
+ objectStorageForcePathStyle: optional("OPENGENI_OBJECT_STORAGE_FORCE_PATH_STYLE"),
1666
+ objectStorageAzureConnectionString: optional("OPENGENI_OBJECT_STORAGE_AZURE_CONNECTION_STRING"),
1667
+ objectStorageAzureAccountName: optional("OPENGENI_OBJECT_STORAGE_AZURE_ACCOUNT_NAME"),
1668
+ objectStorageAzureAccountKey: optional("OPENGENI_OBJECT_STORAGE_AZURE_ACCOUNT_KEY"),
1669
+ objectStorageAzureEndpoint: optional("OPENGENI_OBJECT_STORAGE_AZURE_ENDPOINT"),
1670
+ objectStorageGcsProjectId: optional("OPENGENI_OBJECT_STORAGE_GCS_PROJECT_ID"),
1671
+ objectStorageGcsCredentialsJson: optional("OPENGENI_OBJECT_STORAGE_GCS_CREDENTIALS_JSON"),
1672
+ objectStorageGcsKeyFilename: optional("OPENGENI_OBJECT_STORAGE_GCS_KEY_FILENAME"),
1673
+ objectStorageGcsApiEndpoint: optional("OPENGENI_OBJECT_STORAGE_GCS_API_ENDPOINT")
1674
+ });
1675
+ dbSearchPath(settings);
1676
+ validateObjectStorageSettings(settings);
1677
+ return settings;
1678
+ }
1158
1679
  function configuredGoogleDriveSyncLimits(settings) {
1159
1680
  return KnowledgeSourceSyncLimits.parse({
1160
1681
  maxItems: settings.googleDriveSyncMaxItems,
@@ -1195,6 +1716,24 @@ function personalGitHubOAuthCallbackUrl(publicBaseUrl) {
1195
1716
  const origin = canonicalPublicOrigin(publicBaseUrl);
1196
1717
  return origin ? `${origin}/v1/integrations/github-personal/oauth/callback` : null;
1197
1718
  }
1719
+ function usableDeploymentSecret(value) {
1720
+ return isUsableVoiceInputSecret(value) ? value : void 0;
1721
+ }
1722
+ function usableJevApiKey(settings) {
1723
+ return usableDeploymentSecret(settings.jevApiKey);
1724
+ }
1725
+ function agentConfigDeploymentPolicy(settings) {
1726
+ return agentConfigDeploymentLimitsFromAllowlist(
1727
+ resolveFirstPartyMcpToolPolicy(settings).allowed,
1728
+ settings.webSearchEnabled ? {} : { webSearch: "web search is turned off on this server" }
1729
+ );
1730
+ }
1731
+ function codeSearchDeploymentPolicy(settings) {
1732
+ const available = settings.codeSearchMode !== "off" && usableJevApiKey(settings) !== void 0;
1733
+ if (!available) return { available: false, workspaceDefault: "off" };
1734
+ const workspaceDefault = settings.codeSearchMode === "default_on" ? "on" : settings.codeSearchMode === "experiment" ? "split" : "off";
1735
+ return { available, workspaceDefault };
1736
+ }
1198
1737
  function isUsableVoiceInputSecret(value) {
1199
1738
  if (value == null) return false;
1200
1739
  const trimmed = value.trim();
@@ -1279,18 +1818,18 @@ function voiceInputDeploymentConfigured(settings) {
1279
1818
  (provider) => provider.kind !== "codex-subscription" && provider.kind !== "supergrok-subscription"
1280
1819
  );
1281
1820
  }
1282
- var ModelPricingSchema = z.object({
1283
- inputMicrosPerMillionTokens: z.number().int().nonnegative(),
1284
- cachedInputMicrosPerMillionTokens: z.number().int().nonnegative().optional(),
1285
- cacheWriteMicrosPerMillionTokens: z.number().int().nonnegative().optional(),
1286
- outputMicrosPerMillionTokens: z.number().int().nonnegative(),
1287
- marginBps: z.number().int().min(0).max(1e5).optional()
1821
+ var ModelPricingSchema = z2.object({
1822
+ inputMicrosPerMillionTokens: z2.number().int().nonnegative(),
1823
+ cachedInputMicrosPerMillionTokens: z2.number().int().nonnegative().optional(),
1824
+ cacheWriteMicrosPerMillionTokens: z2.number().int().nonnegative().optional(),
1825
+ outputMicrosPerMillionTokens: z2.number().int().nonnegative(),
1826
+ marginBps: z2.number().int().min(0).max(1e5).optional()
1288
1827
  });
1289
- var ModelPricingScheduleSchema = z.object({
1828
+ var ModelPricingScheduleSchema = z2.object({
1290
1829
  default: ModelPricingSchema,
1291
- inputTokenTiers: z.array(
1292
- z.object({
1293
- minimumInputTokens: z.number().int().nonnegative(),
1830
+ inputTokenTiers: z2.array(
1831
+ z2.object({
1832
+ minimumInputTokens: z2.number().int().nonnegative(),
1294
1833
  pricing: ModelPricingSchema
1295
1834
  })
1296
1835
  ).optional()
@@ -1307,10 +1846,10 @@ var ModelPricingScheduleSchema = z.object({
1307
1846
  previous = tier.minimumInputTokens;
1308
1847
  }
1309
1848
  });
1310
- var CapabilitySupportV1 = z.enum(["supported", "unsupported", "unknown"]);
1311
- var CapabilityStateV1Schema = z.object({
1849
+ var CapabilitySupportV1 = z2.enum(["supported", "unsupported", "unknown"]);
1850
+ var CapabilityStateV1Schema = z2.object({
1312
1851
  upstream: CapabilitySupportV1,
1313
- runnable: z.boolean()
1852
+ runnable: z2.boolean()
1314
1853
  }).superRefine((state, ctx) => {
1315
1854
  if (state.upstream === "unsupported" && state.runnable) {
1316
1855
  ctx.addIssue({
@@ -1320,17 +1859,17 @@ var CapabilityStateV1Schema = z.object({
1320
1859
  });
1321
1860
  }
1322
1861
  });
1323
- var ModelModalityV1 = z.enum(["text", "image", "audio"]);
1324
- var ModelLatencyModeV1 = z.enum(["standard", "priority", "fast"]);
1325
- var ModelCapabilitiesV1Schema = z.object({
1862
+ var ModelModalityV1 = z2.enum(["text", "image", "audio"]);
1863
+ var ModelLatencyModeV1 = z2.enum(["standard", "priority", "fast"]);
1864
+ var ModelCapabilitiesV1Schema = z2.object({
1326
1865
  reasoning: CapabilityStateV1Schema.extend({
1327
- efforts: z.array(ReasoningEffort),
1866
+ efforts: z2.array(ReasoningEffort),
1328
1867
  defaultEffort: ReasoningEffort.nullable(),
1329
- required: z.boolean()
1868
+ required: z2.boolean()
1330
1869
  }),
1331
1870
  functionCalling: CapabilityStateV1Schema,
1332
1871
  structuredOutput: CapabilityStateV1Schema,
1333
- hostedTools: z.object({
1872
+ hostedTools: z2.object({
1334
1873
  webSearch: CapabilityStateV1Schema,
1335
1874
  xSearch: CapabilityStateV1Schema,
1336
1875
  codeExecution: CapabilityStateV1Schema,
@@ -1339,24 +1878,24 @@ var ModelCapabilitiesV1Schema = z.object({
1339
1878
  runnable: false
1340
1879
  })
1341
1880
  }),
1342
- inputModalities: z.array(ModelModalityV1).min(1),
1881
+ inputModalities: z2.array(ModelModalityV1).min(1),
1343
1882
  /** Exact MIME types accepted as typed `input_file`; `text/*` is allowed. */
1344
- inputFileMediaTypes: z.array(z.string()).default([]),
1345
- outputModalities: z.array(ModelModalityV1).min(1),
1346
- transports: z.object({
1883
+ inputFileMediaTypes: z2.array(z2.string()).default([]),
1884
+ outputModalities: z2.array(ModelModalityV1).min(1),
1885
+ transports: z2.object({
1347
1886
  sse: CapabilityStateV1Schema,
1348
1887
  responsesWebSocket: CapabilityStateV1Schema,
1349
1888
  realtimeAudio: CapabilityStateV1Schema
1350
1889
  }),
1351
1890
  promptCaching: CapabilityStateV1Schema.extend({
1352
- mode: z.enum(["implicit", "automatic", "none"])
1891
+ mode: z2.enum(["implicit", "automatic", "none"])
1353
1892
  }).optional(),
1354
- latencyModes: z.array(
1355
- z.object({
1893
+ latencyModes: z2.array(
1894
+ z2.object({
1356
1895
  id: ModelLatencyModeV1,
1357
1896
  upstream: CapabilitySupportV1,
1358
- runnable: z.boolean(),
1359
- billingMultiplierBps: z.number().int().positive().optional()
1897
+ runnable: z2.boolean(),
1898
+ billingMultiplierBps: z2.number().int().positive().optional()
1360
1899
  })
1361
1900
  ).min(1)
1362
1901
  }).superRefine((capabilities, ctx) => {
@@ -1410,9 +1949,9 @@ var ModelCapabilitiesV1Schema = z.object({
1410
1949
  }
1411
1950
  }
1412
1951
  });
1413
- var ModelProviderApi = z.enum(["responses", "chat"]);
1414
- var ModelProviderWireProfile = z.enum(["openai", "azure-openai"]);
1415
- var RegistryProviderKind = z.enum([
1952
+ var ModelProviderApi = z2.enum(["responses", "chat", "anthropic-messages"]);
1953
+ var ModelProviderWireProfile = z2.enum(["openai", "azure-openai"]);
1954
+ var RegistryProviderKind = z2.enum([
1416
1955
  "api-key",
1417
1956
  "anonymous",
1418
1957
  "codex-subscription",
@@ -1420,36 +1959,42 @@ var RegistryProviderKind = z.enum([
1420
1959
  "vercel-gateway-managed",
1421
1960
  "vercel-gateway-workspace",
1422
1961
  "vercel-gateway-organization",
1962
+ "direct-openai-workspace",
1963
+ "direct-azure-workspace",
1423
1964
  "openrouter-workspace",
1424
- "openrouter-organization"
1965
+ "openrouter-organization",
1966
+ "anthropic-organization",
1967
+ "claude-subscription-organization",
1968
+ "anthropic-workspace",
1969
+ "claude-subscription-workspace"
1425
1970
  ]);
1426
- var RegistryModelSchema = z.object({
1427
- id: z.string().min(1),
1971
+ var RegistryModelSchema = z2.object({
1972
+ id: z2.string().min(1),
1428
1973
  // canonical OpenGeni product id
1429
- upstreamModelId: z.string().min(1).optional(),
1974
+ upstreamModelId: z2.string().min(1).optional(),
1430
1975
  // exact provider slug; defaults to id
1431
- aliases: z.array(z.string().min(1)).optional(),
1976
+ aliases: z2.array(z2.string().min(1)).optional(),
1432
1977
  // accepted input only; never sent upstream
1433
- label: z.string().min(1).optional(),
1978
+ label: z2.string().min(1).optional(),
1434
1979
  // display name; defaults to id
1435
- shortLabel: z.string().min(1).max(64).optional(),
1980
+ shortLabel: z2.string().min(1).max(64).optional(),
1436
1981
  // compact UI label; optional
1437
- contextWindowTokens: z.number().int().positive().optional(),
1438
- effectiveContextWindowTokens: z.number().int().positive().optional(),
1439
- autoCompactTokenLimit: z.number().int().positive().optional(),
1982
+ contextWindowTokens: z2.number().int().positive().optional(),
1983
+ effectiveContextWindowTokens: z2.number().int().positive().optional(),
1984
+ autoCompactTokenLimit: z2.number().int().positive().optional(),
1440
1985
  // Canonical model-facing function/tool-result policy. The runtime applies
1441
1986
  // the same 1.2x serialization allowance as Codex when materializing output.
1442
- toolOutputTruncationTokens: z.number().int().positive().optional(),
1443
- reasoningEffort: z.boolean().optional(),
1987
+ toolOutputTruncationTokens: z2.number().int().positive().optional(),
1988
+ reasoningEffort: z2.boolean().optional(),
1444
1989
  // legacy compatibility input/projection
1445
- hostedWebSearch: z.boolean().optional(),
1990
+ hostedWebSearch: z2.boolean().optional(),
1446
1991
  // legacy compatibility input/projection
1447
1992
  capabilities: ModelCapabilitiesV1Schema.optional(),
1448
- pricing: z.union([ModelPricingSchema, ModelPricingScheduleSchema]).optional(),
1993
+ pricing: z2.union([ModelPricingSchema, ModelPricingScheduleSchema]).optional(),
1449
1994
  // Reserved normalized contracts are derived by OpenGeni in V1. Generic
1450
1995
  // registry JSON must not opt itself into workspace BYOK or reattribute cost.
1451
- credentialSource: z.never().optional(),
1452
- billing: z.never().optional()
1996
+ credentialSource: z2.never().optional(),
1997
+ billing: z2.never().optional()
1453
1998
  }).superRefine((model, ctx) => {
1454
1999
  if (model.capabilities && model.reasoningEffort !== void 0 && model.reasoningEffort !== model.capabilities.reasoning.runnable) {
1455
2000
  ctx.addIssue({
@@ -1466,27 +2011,54 @@ var RegistryModelSchema = z.object({
1466
2011
  });
1467
2012
  }
1468
2013
  });
1469
- var RegistryProviderSchema = z.object({
2014
+ var ClaudeSubscriptionIdentity = z2.object({
2015
+ accountUuid: z2.union([z2.string().uuid(), z2.literal("")]),
2016
+ deviceId: z2.string().regex(/^[a-f0-9]{64}$/)
2017
+ }).strict();
2018
+ var ClaudeSubscriptionCredential = z2.object({
2019
+ version: z2.literal(1),
2020
+ token: z2.string().regex(/^sk-ant-oat[0-9]+-\S+$/),
2021
+ identity: ClaudeSubscriptionIdentity,
2022
+ oauth: z2.object({
2023
+ refreshToken: z2.string().min(1).max(16384),
2024
+ expiresAt: z2.string().datetime(),
2025
+ scopes: z2.array(z2.string().min(1).max(256)).min(1).max(32)
2026
+ }).strict().optional()
2027
+ }).strict();
2028
+ var AnthropicProviderOptions = z2.object({
2029
+ identity: ClaudeSubscriptionIdentity.optional(),
2030
+ // Native connection provenance only; never included in Anthropic request bodies.
2031
+ credentialBinding: z2.object({
2032
+ connectionId: z2.string().uuid(),
2033
+ credentialVersion: z2.number().int().positive()
2034
+ }).strict().optional(),
2035
+ auth: z2.enum(["api-key", "oauth"]).default("api-key"),
2036
+ cacheTtl: z2.enum(["5m", "1h", "off"]).default("5m"),
2037
+ maxOutputTokens: z2.number().int().positive().default(32e3),
2038
+ streamIdleTimeoutMs: z2.number().int().positive().default(6e5)
2039
+ });
2040
+ var RegistryProviderSchema = z2.object({
1470
2041
  kind: RegistryProviderKind.default("api-key"),
1471
- id: z.string().min(1).regex(registryId),
2042
+ id: z2.string().min(1).regex(registryId),
1472
2043
  // stable provider id, e.g. "fireworks"
1473
- label: z.string().min(1).optional(),
2044
+ label: z2.string().min(1).optional(),
1474
2045
  api: ModelProviderApi.default("chat"),
2046
+ anthropic: AnthropicProviderOptions.optional(),
1475
2047
  wireProfile: ModelProviderWireProfile.default("openai"),
1476
- baseUrl: z.string().url(),
1477
- apiKey: z.string().optional(),
2048
+ baseUrl: z2.string().url(),
2049
+ apiKey: z2.string().optional(),
1478
2050
  // inline key (pragmatic) ...
1479
- apiKeyEnv: z.string().optional(),
2051
+ apiKeyEnv: z2.string().optional(),
1480
2052
  // ... OR name of the env var holding the key (preferred)
1481
- defaultQuery: z.record(z.string(), z.string()).optional(),
1482
- defaultHeaders: z.record(z.string(), z.string()).optional(),
1483
- publicDefaultQueryNames: z.array(z.string().min(1)).optional(),
1484
- publicDefaultHeaderNames: z.array(z.string().min(1)).optional(),
2053
+ defaultQuery: z2.record(z2.string(), z2.string()).optional(),
2054
+ defaultHeaders: z2.record(z2.string(), z2.string()).optional(),
2055
+ publicDefaultQueryNames: z2.array(z2.string().min(1)).optional(),
2056
+ publicDefaultHeaderNames: z2.array(z2.string().min(1)).optional(),
1485
2057
  // V1 derives these from provider kind. Workspace BYOK is deliberately not a
1486
2058
  // registry switch and requires a separately reviewed encrypted broker.
1487
- credentialSource: z.never().optional(),
1488
- billing: z.never().optional(),
1489
- models: z.array(RegistryModelSchema).min(1)
2059
+ credentialSource: z2.never().optional(),
2060
+ billing: z2.never().optional(),
2061
+ models: z2.array(RegistryModelSchema).min(1)
1490
2062
  }).superRefine((provider, ctx) => {
1491
2063
  if (provider.kind !== "anonymous") {
1492
2064
  return;
@@ -1556,17 +2128,21 @@ var RESERVED_MODEL_PROVIDER_IDS = /* @__PURE__ */ new Set([
1556
2128
  ORGANIZATION_GATEWAY_PROVIDER_ID,
1557
2129
  OPENROUTER_PROVIDER_ID,
1558
2130
  WORKSPACE_OPENROUTER_PROVIDER_ID,
1559
- ORGANIZATION_OPENROUTER_PROVIDER_ID
2131
+ ORGANIZATION_OPENROUTER_PROVIDER_ID,
2132
+ "organization-anthropic",
2133
+ "organization-claude-subscription",
2134
+ "workspace-anthropic",
2135
+ "workspace-claude-subscription"
1560
2136
  ]);
1561
- var ModelCostClass = z.enum(["free", "credits"]);
1562
- var ConfiguredModelCostClass = z.enum([
2137
+ var ModelCostClass = z2.enum(["free", "credits"]);
2138
+ var ConfiguredModelCostClass = z2.enum([
1563
2139
  "free",
1564
2140
  "credits",
1565
2141
  "subscription",
1566
2142
  "workspace",
1567
2143
  "organization"
1568
2144
  ]);
1569
- var ModelNote = z.string().max(500).refine((value) => !/[\r\n|]/u.test(value), {
2145
+ var ModelNote = z2.string().max(500).refine((value) => !/[\r\n|]/u.test(value), {
1570
2146
  message: "model notes must not contain newlines or the | field separator"
1571
2147
  });
1572
2148
  function parseModelCostPolicyJson(raw) {
@@ -1579,7 +2155,7 @@ function parseModelCostPolicyJson(raw) {
1579
2155
  { cause: error }
1580
2156
  );
1581
2157
  }
1582
- return z.record(z.string().min(1), ModelCostClass).parse(parsed);
2158
+ return z2.record(z2.string().min(1), ModelCostClass).parse(parsed);
1583
2159
  }
1584
2160
  function parseModelNotesJson(raw) {
1585
2161
  let parsed;
@@ -1591,45 +2167,45 @@ function parseModelNotesJson(raw) {
1591
2167
  { cause: error }
1592
2168
  );
1593
2169
  }
1594
- return z.record(z.string().min(1), ModelNote).parse(parsed);
2170
+ return z2.record(z2.string().min(1), ModelNote).parse(parsed);
1595
2171
  }
1596
2172
  function configuredModelNotes(settings) {
1597
2173
  return parseModelNotesJson(settings.modelNotesJson);
1598
2174
  }
1599
- var GatewayCatalogModel = z.object({
1600
- productId: z.string().min(1),
1601
- workspaceProductId: z.string().min(1).startsWith(WORKSPACE_GATEWAY_MODEL_ID_PREFIX),
1602
- upstreamModelId: z.string().min(1),
1603
- label: z.string().min(1),
1604
- shortLabel: z.string().min(1).max(64).optional(),
1605
- providers: z.array(z.string().min(1)).min(1),
1606
- implicitCaching: z.boolean().default(false),
1607
- vision: z.boolean().default(false),
1608
- inputFileMediaTypes: z.array(z.string().min(1)).default([]),
1609
- contextWindowTokens: z.number().int().positive().default(1e6),
1610
- effectiveContextWindowTokens: z.number().int().positive().default(9e5),
1611
- autoCompactTokenLimit: z.number().int().positive().default(85e4),
1612
- pricing: z.union([ModelPricingSchema, ModelPricingScheduleSchema]).optional(),
1613
- credentialSource: z.never().optional(),
1614
- billing: z.never().optional(),
1615
- apiKey: z.never().optional()
2175
+ var GatewayCatalogModel = z2.object({
2176
+ productId: z2.string().min(1),
2177
+ workspaceProductId: z2.string().min(1).startsWith(WORKSPACE_GATEWAY_MODEL_ID_PREFIX),
2178
+ upstreamModelId: z2.string().min(1),
2179
+ label: z2.string().min(1),
2180
+ shortLabel: z2.string().min(1).max(64).optional(),
2181
+ providers: z2.array(z2.string().min(1)).min(1),
2182
+ implicitCaching: z2.boolean().default(false),
2183
+ vision: z2.boolean().default(false),
2184
+ inputFileMediaTypes: z2.array(z2.string().min(1)).default([]),
2185
+ contextWindowTokens: z2.number().int().positive().default(1e6),
2186
+ effectiveContextWindowTokens: z2.number().int().positive().default(9e5),
2187
+ autoCompactTokenLimit: z2.number().int().positive().default(85e4),
2188
+ pricing: z2.union([ModelPricingSchema, ModelPricingScheduleSchema]).optional(),
2189
+ credentialSource: z2.never().optional(),
2190
+ billing: z2.never().optional(),
2191
+ apiKey: z2.never().optional()
1616
2192
  }).strict();
1617
- var OpenRouterCatalogModel = z.object({
1618
- upstreamModelId: z.string().min(1).endsWith(":free"),
1619
- label: z.string().min(1),
1620
- shortLabel: z.string().min(1).max(64).optional(),
1621
- aliases: z.array(z.string().min(1)).default([]),
2193
+ var OpenRouterCatalogModel = z2.object({
2194
+ upstreamModelId: z2.string().min(1).endsWith(":free"),
2195
+ label: z2.string().min(1),
2196
+ shortLabel: z2.string().min(1).max(64).optional(),
2197
+ aliases: z2.array(z2.string().min(1)).default([]),
1622
2198
  capabilities: ModelCapabilitiesV1Schema,
1623
- contextWindowTokens: z.number().int().positive().optional(),
1624
- effectiveContextWindowTokens: z.number().int().positive().optional(),
1625
- autoCompactTokenLimit: z.number().int().positive().optional(),
1626
- toolOutputTruncationTokens: z.number().int().positive().optional(),
1627
- credentialSource: z.never().optional(),
1628
- billing: z.never().optional(),
1629
- pricing: z.never().optional(),
1630
- apiKey: z.never().optional()
2199
+ contextWindowTokens: z2.number().int().positive().optional(),
2200
+ effectiveContextWindowTokens: z2.number().int().positive().optional(),
2201
+ autoCompactTokenLimit: z2.number().int().positive().optional(),
2202
+ toolOutputTruncationTokens: z2.number().int().positive().optional(),
2203
+ credentialSource: z2.never().optional(),
2204
+ billing: z2.never().optional(),
2205
+ pricing: z2.never().optional(),
2206
+ apiKey: z2.never().optional()
1631
2207
  }).strict();
1632
- var DeploymentRegistryBaseUrl = z.string().url().superRefine((value, context) => {
2208
+ var DeploymentRegistryBaseUrl = z2.string().url().superRefine((value, context) => {
1633
2209
  const url = new URL(value);
1634
2210
  if (url.username || url.password) {
1635
2211
  context.addIssue({
@@ -1650,40 +2226,66 @@ var DeploymentRegistryBaseUrl = z.string().url().superRefine((value, context) =>
1650
2226
  });
1651
2227
  }
1652
2228
  });
1653
- var DeploymentRegistryProviderKind = z.enum(["api-key", "anonymous"]);
2229
+ var DeploymentRegistryProviderKind = z2.enum(["api-key", "anonymous"]);
1654
2230
  var DeploymentRegistryModelSchema = RegistryModelSchema.safeExtend({
1655
- pricing: z.never().optional()
2231
+ pricing: z2.never().optional()
1656
2232
  }).strict();
1657
2233
  var DeploymentRegistryProviderSchema = RegistryProviderSchema.safeExtend({
1658
2234
  kind: DeploymentRegistryProviderKind.default("api-key"),
1659
2235
  baseUrl: DeploymentRegistryBaseUrl,
1660
- models: z.array(DeploymentRegistryModelSchema).min(1),
1661
- apiKey: z.never().optional(),
1662
- apiKeyEnv: z.never().optional(),
1663
- defaultHeaders: z.never().optional(),
1664
- defaultQuery: z.never().optional(),
1665
- publicDefaultHeaderNames: z.never().optional(),
1666
- publicDefaultQueryNames: z.never().optional()
2236
+ models: z2.array(DeploymentRegistryModelSchema).min(1),
2237
+ apiKey: z2.never().optional(),
2238
+ apiKeyEnv: z2.never().optional(),
2239
+ defaultHeaders: z2.never().optional(),
2240
+ defaultQuery: z2.never().optional(),
2241
+ publicDefaultHeaderNames: z2.never().optional(),
2242
+ publicDefaultQueryNames: z2.never().optional()
1667
2243
  }).strict();
1668
2244
  var DeploymentGatewayCatalogModelSchema = GatewayCatalogModel.safeExtend({
1669
- pricing: z.never().optional()
2245
+ pricing: z2.never().optional()
1670
2246
  }).strict();
1671
- var ModelCatalogDocument = z.object({
1672
- schemaVersion: z.literal(1),
2247
+ var CodexCatalogModelSchema = RegistryModelSchema.safeExtend({
2248
+ retired: z2.boolean().optional(),
2249
+ id: z2.string().regex(/^codex\/[a-zA-Z0-9][a-zA-Z0-9._-]*$/),
2250
+ upstreamModelId: z2.string().regex(/^[a-zA-Z0-9][a-zA-Z0-9._-]*$/),
2251
+ capabilities: ModelCapabilitiesV1Schema,
2252
+ pricing: z2.never().optional()
2253
+ }).strict().superRefine((model, context) => {
2254
+ if (model.id !== `${CODEX_MODEL_ID_PREFIX}${model.upstreamModelId}`) {
2255
+ context.addIssue({
2256
+ code: "custom",
2257
+ path: ["id"],
2258
+ message: "Codex product id must match its upstream model slug"
2259
+ });
2260
+ }
2261
+ }).transform((model) => ({
2262
+ ...model,
2263
+ capabilities: reviewedCodexCatalogCapabilities(model.upstreamModelId, model.capabilities)
2264
+ }));
2265
+ var ModelCatalogDocument = z2.object({
2266
+ schemaVersion: z2.literal(1),
1673
2267
  /** Canonical deployment default. Omission preserves the V1 first-built-in
1674
2268
  * fallback for existing documents; operators should set this explicitly
1675
2269
  * when cutting over a registry or connected-subscription default. */
1676
- defaultModel: z.string().min(1).optional(),
1677
- builtInModels: z.array(z.string().min(1)).min(1),
1678
- registryProviders: z.array(DeploymentRegistryProviderSchema).default([]),
1679
- gatewayModels: z.array(DeploymentGatewayCatalogModelSchema).default([]),
1680
- openrouterModels: z.array(OpenRouterCatalogModel).default([]),
1681
- modelNotes: z.record(z.string().min(1), ModelNote).default({}),
1682
- billing: z.never().optional(),
1683
- enabled: z.never().optional(),
1684
- apiKey: z.never().optional(),
1685
- bands: z.never().optional()
2270
+ defaultModel: z2.string().min(1).optional(),
2271
+ builtInModels: z2.array(z2.string().min(1)),
2272
+ registryProviders: z2.array(DeploymentRegistryProviderSchema).default([]),
2273
+ codexModels: z2.array(CodexCatalogModelSchema).optional(),
2274
+ gatewayModels: z2.array(DeploymentGatewayCatalogModelSchema).default([]),
2275
+ openrouterModels: z2.array(OpenRouterCatalogModel).default([]),
2276
+ modelNotes: z2.record(z2.string().min(1), ModelNote).default({}),
2277
+ billing: z2.never().optional(),
2278
+ enabled: z2.never().optional(),
2279
+ apiKey: z2.never().optional(),
2280
+ bands: z2.never().optional()
1686
2281
  }).strict().superRefine((document, context) => {
2282
+ if (document.builtInModels.length === 0 && !document.defaultModel) {
2283
+ context.addIssue({
2284
+ code: "custom",
2285
+ path: ["defaultModel"],
2286
+ message: "catalogs without built-in models require an explicit default model"
2287
+ });
2288
+ }
1687
2289
  const productIds = /* @__PURE__ */ new Set();
1688
2290
  const providerIds = /* @__PURE__ */ new Set();
1689
2291
  const gatewayUpstreamIds = /* @__PURE__ */ new Set();
@@ -1705,8 +2307,16 @@ var ModelCatalogDocument = z.object({
1705
2307
  productIds.add(id);
1706
2308
  };
1707
2309
  document.builtInModels.forEach((id, index) => add(id, ["builtInModels", index]));
2310
+ document.codexModels?.forEach((model, index) => add(model.id, ["codexModels", index, "id"]));
2311
+ if (document.codexModels?.some((model) => model.retired && model.id === document.defaultModel)) {
2312
+ context.addIssue({
2313
+ code: "custom",
2314
+ path: ["defaultModel"],
2315
+ message: "a retired model cannot be the deployment default"
2316
+ });
2317
+ }
1708
2318
  document.registryProviders.forEach((provider, providerIndex) => {
1709
- if (RESERVED_MODEL_PROVIDER_IDS.has(provider.id)) {
2319
+ if (RESERVED_MODEL_PROVIDER_IDS.has(provider.id) || provider.id.startsWith("workspace-openai-") || provider.id.startsWith("workspace-azure-openai-")) {
1710
2320
  context.addIssue({
1711
2321
  code: "custom",
1712
2322
  path: ["registryProviders", providerIndex, "id"],
@@ -1751,7 +2361,7 @@ var ModelCatalogDocument = z.object({
1751
2361
  message: "catalog default model must not contain newlines or the | field separator"
1752
2362
  });
1753
2363
  }
1754
- if (document.defaultModel && !productIds.has(document.defaultModel) && !document.defaultModel.startsWith(CODEX_MODEL_ID_PREFIX) && !document.defaultModel.startsWith(XAI_SUBSCRIPTION_MODEL_ID_PREFIX)) {
2364
+ if (document.defaultModel && !productIds.has(document.defaultModel) && !(document.codexModels === void 0 && document.defaultModel.startsWith(CODEX_MODEL_ID_PREFIX)) && !document.defaultModel.startsWith(XAI_SUBSCRIPTION_MODEL_ID_PREFIX)) {
1755
2365
  context.addIssue({
1756
2366
  code: "custom",
1757
2367
  path: ["defaultModel"],
@@ -1821,15 +2431,19 @@ function applyModelCatalogDocument(settings, rawDocument) {
1821
2431
  ),
1822
2432
  resolvedGatewayModelsJson: JSON.stringify(document.gatewayModels),
1823
2433
  resolvedOpenRouterModelsJson: JSON.stringify(document.openrouterModels),
2434
+ resolvedCodexModelsJson: document.codexModels === void 0 ? void 0 : JSON.stringify(document.codexModels),
1824
2435
  modelNotesJson: JSON.stringify(document.modelNotes)
1825
2436
  };
1826
2437
  return resolved;
1827
2438
  }
1828
- var IntegrationOAuthClientConfigSchema = z.object({
1829
- clientId: z.string().min(1),
1830
- clientSecret: z.string().min(1).optional(),
1831
- tokenEndpointAuthMethod: z.enum(["none", "client_secret_post", "client_secret_basic"]).default("none")
2439
+ var IntegrationOAuthClientConfigSchema = z2.object({
2440
+ clientId: z2.string().min(1),
2441
+ clientSecret: z2.string().min(1).optional(),
2442
+ tokenEndpointAuthMethod: z2.enum(["none", "client_secret_post", "client_secret_basic"]).default("none")
1832
2443
  });
2444
+ function isManagedOpenRouterFreeRoute(route) {
2445
+ return route.provider.kind === "openrouter-managed" && route.configured.upstreamModelId.endsWith(":free");
2446
+ }
1833
2447
  var VERCEL_AI_GATEWAY_BASE_URL = "https://ai-gateway.vercel.sh/v1";
1834
2448
  var VERCEL_AI_GATEWAY_AI_SDK_BASE_URL = "https://ai-gateway.vercel.sh/v4/ai";
1835
2449
  var VERCEL_AI_GATEWAY_CONNECTION_DOMAIN = "ai-gateway.vercel.sh";
@@ -1956,7 +2570,7 @@ function configuredGatewayCatalogModels(settings) {
1956
2570
  if (settings.resolvedGatewayModelsJson === void 0) {
1957
2571
  return defaultGatewayCatalogModels();
1958
2572
  }
1959
- return z.array(GatewayCatalogModel).parse(JSON.parse(settings.resolvedGatewayModelsJson));
2573
+ return z2.array(GatewayCatalogModel).parse(JSON.parse(settings.resolvedGatewayModelsJson));
1960
2574
  }
1961
2575
  function configuredGatewayUpstreamModelIds(settings) {
1962
2576
  return configuredGatewayCatalogModels(settings).map((model) => model.upstreamModelId);
@@ -1975,7 +2589,7 @@ function configuredOpenRouterCatalogModels(settings) {
1975
2589
  if (settings.resolvedOpenRouterModelsJson === void 0) {
1976
2590
  return [...OPENGENI_OPENROUTER_MODELS];
1977
2591
  }
1978
- return z.array(OpenRouterCatalogModel).parse(JSON.parse(settings.resolvedOpenRouterModelsJson));
2592
+ return z2.array(OpenRouterCatalogModel).parse(JSON.parse(settings.resolvedOpenRouterModelsJson));
1979
2593
  }
1980
2594
  function configuredOpenRouterUpstreamModelIds(settings) {
1981
2595
  return configuredOpenRouterCatalogModels(settings).map((model) => model.upstreamModelId);
@@ -1993,6 +2607,73 @@ function configuredOpenRouterOrganizationProductModelIds(settings) {
1993
2607
  return [];
1994
2608
  }
1995
2609
  var defaultModelPricing = {
2610
+ "gpt-6-astra": {
2611
+ default: {
2612
+ // OpenAI list price: $10 / $1 cached / $12.50 cache write / $50 output.
2613
+ inputMicrosPerMillionTokens: 1e7,
2614
+ cachedInputMicrosPerMillionTokens: 1e6,
2615
+ cacheWriteMicrosPerMillionTokens: 125e5,
2616
+ outputMicrosPerMillionTokens: 5e7,
2617
+ marginBps: 500
2618
+ },
2619
+ inputTokenTiers: [
2620
+ {
2621
+ // Prompts with more than 272K input tokens use 2x input and 1.5x output.
2622
+ minimumInputTokens: 272001,
2623
+ pricing: {
2624
+ inputMicrosPerMillionTokens: 2e7,
2625
+ cachedInputMicrosPerMillionTokens: 2e6,
2626
+ cacheWriteMicrosPerMillionTokens: 25e6,
2627
+ outputMicrosPerMillionTokens: 75e6,
2628
+ marginBps: 500
2629
+ }
2630
+ }
2631
+ ]
2632
+ },
2633
+ "gpt-6-sol": {
2634
+ default: {
2635
+ // OpenAI list price: $2 / $0.20 cached / $2.50 cache write / $10 output.
2636
+ inputMicrosPerMillionTokens: 2e6,
2637
+ cachedInputMicrosPerMillionTokens: 2e5,
2638
+ cacheWriteMicrosPerMillionTokens: 25e5,
2639
+ outputMicrosPerMillionTokens: 1e7,
2640
+ marginBps: 500
2641
+ },
2642
+ inputTokenTiers: [
2643
+ {
2644
+ minimumInputTokens: 272001,
2645
+ pricing: {
2646
+ inputMicrosPerMillionTokens: 4e6,
2647
+ cachedInputMicrosPerMillionTokens: 4e5,
2648
+ cacheWriteMicrosPerMillionTokens: 5e6,
2649
+ outputMicrosPerMillionTokens: 15e6,
2650
+ marginBps: 500
2651
+ }
2652
+ }
2653
+ ]
2654
+ },
2655
+ "gpt-6-luna": {
2656
+ default: {
2657
+ // OpenAI list price: $0.10 / $0.01 cached / $0.125 cache write / $0.50 output.
2658
+ inputMicrosPerMillionTokens: 1e5,
2659
+ cachedInputMicrosPerMillionTokens: 1e4,
2660
+ cacheWriteMicrosPerMillionTokens: 125e3,
2661
+ outputMicrosPerMillionTokens: 5e5,
2662
+ marginBps: 500
2663
+ },
2664
+ inputTokenTiers: [
2665
+ {
2666
+ minimumInputTokens: 272001,
2667
+ pricing: {
2668
+ inputMicrosPerMillionTokens: 2e5,
2669
+ cachedInputMicrosPerMillionTokens: 2e4,
2670
+ cacheWriteMicrosPerMillionTokens: 25e4,
2671
+ outputMicrosPerMillionTokens: 75e4,
2672
+ marginBps: 500
2673
+ }
2674
+ }
2675
+ ]
2676
+ },
1996
2677
  "gpt-5.6-sol": {
1997
2678
  default: {
1998
2679
  // Promotional OpenAI pricing, guaranteed through at least 2026-11-21.
@@ -2092,6 +2773,103 @@ var defaultModelPricing = {
2092
2773
  }
2093
2774
  }
2094
2775
  };
2776
+ var reviewedModelListPricing = {
2777
+ // Reviewed 2026-10-03 against https://developers.openai.com/api/docs/pricing
2778
+ // and https://ai-gateway.vercel.sh/v1/models. GPT-6.1 Sol's cache-read rate
2779
+ // is 5% of input, not GPT-6 Sol's 10%. The long-context boundary is >272K.
2780
+ "gpt-6.1-sol": {
2781
+ default: {
2782
+ inputMicrosPerMillionTokens: 2e6,
2783
+ cachedInputMicrosPerMillionTokens: 1e5,
2784
+ cacheWriteMicrosPerMillionTokens: 25e5,
2785
+ outputMicrosPerMillionTokens: 1e7,
2786
+ marginBps: 500
2787
+ },
2788
+ inputTokenTiers: [
2789
+ {
2790
+ minimumInputTokens: 272001,
2791
+ pricing: {
2792
+ inputMicrosPerMillionTokens: 4e6,
2793
+ cachedInputMicrosPerMillionTokens: 2e5,
2794
+ cacheWriteMicrosPerMillionTokens: 5e6,
2795
+ outputMicrosPerMillionTokens: 15e6,
2796
+ marginBps: 500
2797
+ }
2798
+ }
2799
+ ]
2800
+ },
2801
+ // xAI Standard API list rates, reviewed 2026-10-03:
2802
+ // https://docs.x.ai/developers/models/grok-4.5 (and grok-4.6 / grok-4.7).
2803
+ // Preserve the native API's established >=200K boundary. Gateway's separately
2804
+ // configured >200K schedule is not interchangeable with this native schedule.
2805
+ // xAI's automatic caching has no separate cache-write surcharge.
2806
+ ...Object.fromEntries(
2807
+ [
2808
+ ["grok-4.5", 3e5],
2809
+ ["grok-4.6", 5e5],
2810
+ ["grok-4.7", 5e5]
2811
+ ].map(([model, cachedInput]) => [
2812
+ model,
2813
+ {
2814
+ default: {
2815
+ inputMicrosPerMillionTokens: 2e6,
2816
+ cachedInputMicrosPerMillionTokens: cachedInput,
2817
+ outputMicrosPerMillionTokens: 6e6,
2818
+ marginBps: 500
2819
+ },
2820
+ inputTokenTiers: [
2821
+ {
2822
+ minimumInputTokens: 2e5,
2823
+ pricing: {
2824
+ inputMicrosPerMillionTokens: 4e6,
2825
+ cachedInputMicrosPerMillionTokens: cachedInput * 2,
2826
+ outputMicrosPerMillionTokens: 12e6,
2827
+ marginBps: 500
2828
+ }
2829
+ }
2830
+ ]
2831
+ }
2832
+ ])
2833
+ ),
2834
+ // Every reviewed native Claude profile. Standard/global prices include
2835
+ // 5-minute cache writes; 1-hour native routes are projected separately.
2836
+ // https://platform.claude.com/docs/en/about-claude/pricing (2026-10-03).
2837
+ // Opus 5.5 cache reads are 5%, unlike the older models' 10%; none of these
2838
+ // models has a long-context premium.
2839
+ ...Object.fromEntries(
2840
+ [
2841
+ ["claude-opus-5-5", 4e6, 2e5, 5e6, 2e7],
2842
+ ["claude-sonnet-5-5", 2e6, 2e5, 25e5, 1e7],
2843
+ ["claude-opus-5", 5e6, 5e5, 625e4, 25e6],
2844
+ ["claude-sonnet-5", 2e6, 2e5, 25e5, 1e7],
2845
+ ["claude-opus-4-8", 5e6, 5e5, 625e4, 25e6],
2846
+ ["claude-opus-4-7", 5e6, 5e5, 625e4, 25e6],
2847
+ ["claude-opus-4-6", 5e6, 5e5, 625e4, 25e6],
2848
+ ["claude-sonnet-4-6", 3e6, 3e5, 375e4, 15e6],
2849
+ ["claude-haiku-4-5-20251001", 1e6, 1e5, 125e4, 5e6]
2850
+ ].map(([model, input, cachedInput, cacheWrite, output]) => [
2851
+ model,
2852
+ {
2853
+ default: {
2854
+ inputMicrosPerMillionTokens: input,
2855
+ cachedInputMicrosPerMillionTokens: cachedInput,
2856
+ cacheWriteMicrosPerMillionTokens: cacheWrite,
2857
+ outputMicrosPerMillionTokens: output,
2858
+ marginBps: 500
2859
+ }
2860
+ }
2861
+ ])
2862
+ )
2863
+ };
2864
+ var reviewedFreeOpenRouterListPricing = {
2865
+ default: {
2866
+ inputMicrosPerMillionTokens: 0,
2867
+ cachedInputMicrosPerMillionTokens: 0,
2868
+ cacheWriteMicrosPerMillionTokens: 0,
2869
+ outputMicrosPerMillionTokens: 0,
2870
+ marginBps: 500
2871
+ }
2872
+ };
2095
2873
  var SANDBOX_REQUIRED_ENV = {
2096
2874
  // docker/local/none need no credentials (local dev container / in-process / off).
2097
2875
  docker: [],
@@ -2151,6 +2929,7 @@ function optionalEnvironmentValue(name, source) {
2151
2929
  const value = source[name];
2152
2930
  return value && value.trim().length > 0 ? value : void 0;
2153
2931
  }
2932
+ var idleCommandContainmentDisabledWarned = false;
2154
2933
  function getSettings(source = process.env) {
2155
2934
  const optional = (name) => optionalEnvironmentValue(name, source);
2156
2935
  const modelCatalogSource = optional("OPENGENI_MODEL_CATALOG_SOURCE");
@@ -2161,6 +2940,7 @@ function getSettings(source = process.env) {
2161
2940
  deploymentRevision: optional("OPENGENI_DEPLOYMENT_REVISION") ?? optional("SOURCE_VERSION") ?? optional("GITHUB_SHA"),
2162
2941
  serverVersion: optional("OPENGENI_SERVER_VERSION"),
2163
2942
  databaseUrl: optional("OPENGENI_DATABASE_URL"),
2943
+ apiDatabasePoolMax: optional("OPENGENI_API_DATABASE_POOL_MAX"),
2164
2944
  dbSchema: optional("OPENGENI_DB_SCHEMA"),
2165
2945
  rlsStrategy: optional("OPENGENI_RLS_STRATEGY"),
2166
2946
  runtimeDatabaseRole: optional("OPENGENI_RUNTIME_DATABASE_ROLE"),
@@ -2192,26 +2972,32 @@ function getSettings(source = process.env) {
2192
2972
  observabilityMetricsEnabled: optional("OPENGENI_OBSERVABILITY_METRICS_ENABLED"),
2193
2973
  observabilityOtlpEndpoint: optional("OPENGENI_OTEL_EXPORTER_OTLP_ENDPOINT") ?? optional("OTEL_EXPORTER_OTLP_ENDPOINT"),
2194
2974
  observabilityOtlpHeaders: optional("OPENGENI_OTEL_EXPORTER_OTLP_HEADERS") ?? optional("OTEL_EXPORTER_OTLP_HEADERS"),
2975
+ observabilityDiagnosticsEndpoint: optional("OPENGENI_OBSERVABILITY_DIAGNOSTICS_ENDPOINT"),
2976
+ observabilityDiagnosticsHeaders: optional("OPENGENI_OBSERVABILITY_DIAGNOSTICS_HEADERS"),
2195
2977
  analyticsEnabled: optional("OPENGENI_ANALYTICS_ENABLED"),
2196
2978
  analyticsConsentRequired: optional("OPENGENI_ANALYTICS_CONSENT_REQUIRED"),
2197
2979
  analyticsReoClientId: optional("OPENGENI_ANALYTICS_REO_CLIENT_ID"),
2198
2980
  analyticsPosthogProjectKey: optional("OPENGENI_ANALYTICS_POSTHOG_PROJECT_KEY"),
2199
2981
  analyticsPosthogHost: optional("OPENGENI_ANALYTICS_POSTHOG_HOST"),
2200
2982
  analyticsGa4MeasurementId: optional("OPENGENI_ANALYTICS_GA4_MEASUREMENT_ID"),
2983
+ legalPrivacyPolicyUrl: optional("OPENGENI_LEGAL_PRIVACY_POLICY_URL"),
2984
+ legalTermsOfServiceUrl: optional("OPENGENI_LEGAL_TERMS_OF_SERVICE_URL"),
2985
+ supportEmail: optional("OPENGENI_SUPPORT_EMAIL"),
2201
2986
  publicBaseUrl: optional("OPENGENI_PUBLIC_BASE_URL"),
2987
+ documentationUrl: optional("OPENGENI_DOCUMENTATION_URL"),
2202
2988
  mcpOauthEnabled: optional("OPENGENI_MCP_OAUTH_ENABLED"),
2203
- mcpOauthTrustedProxyHops: optional("OPENGENI_MCP_OAUTH_TRUSTED_PROXY_HOPS"),
2989
+ apiTrustedProxyHops: optional("OPENGENI_API_TRUSTED_PROXY_HOPS"),
2990
+ apiTrustedProxyCidrs: optional("OPENGENI_API_TRUSTED_PROXY_CIDRS"),
2204
2991
  webBaseUrl: optional("OPENGENI_WEB_BASE_URL"),
2205
2992
  agentReleasesBaseUrl: optional("OPENGENI_AGENT_RELEASES_BASE_URL"),
2206
2993
  agentStableVersion: optional("OPENGENI_AGENT_STABLE_VERSION"),
2207
2994
  agentBetaVersion: optional("OPENGENI_AGENT_BETA_VERSION"),
2208
2995
  productAccessMode: optional("OPENGENI_PRODUCT_ACCESS_MODE"),
2209
- organizationTenancyCanonicalActivationEnabled: optional(
2210
- "OPENGENI_ORGANIZATION_TENANCY_CANONICAL_ACTIVATION_ENABLED"
2211
- ),
2212
2996
  billingMode: optional("OPENGENI_BILLING_MODE"),
2997
+ verifiedSignupTrialCreditsEnabled: optional("OPENGENI_VERIFIED_SIGNUP_TRIAL_CREDITS_ENABLED"),
2213
2998
  entitlementsMode: optional("OPENGENI_ENTITLEMENTS_MODE"),
2214
2999
  usageLimitsMode: optional("OPENGENI_USAGE_LIMITS_MODE"),
3000
+ usageAllowancesEnabled: optional("OPENGENI_USAGE_ALLOWANCES_ENABLED"),
2215
3001
  staticEntitlementsJson: optional("OPENGENI_STATIC_ENTITLEMENTS_JSON"),
2216
3002
  staticUsageLimitsJson: optional("OPENGENI_STATIC_USAGE_LIMITS_JSON"),
2217
3003
  delegationSecret: optional("OPENGENI_DELEGATION_SECRET"),
@@ -2231,13 +3017,16 @@ function getSettings(source = process.env) {
2231
3017
  environmentsEncryptionKey: optional("OPENGENI_ENVIRONMENTS_ENCRYPTION_KEY"),
2232
3018
  integrationsEnabled: optional("OPENGENI_INTEGRATIONS_ENABLED"),
2233
3019
  integrationsStateSecret: optional("OPENGENI_INTEGRATIONS_STATE_SECRET"),
3020
+ integrationsOauthShortStateEnabled: optional("OPENGENI_INTEGRATIONS_OAUTH_SHORT_STATE_ENABLED"),
2234
3021
  integrationsAllowPrivateNetworkTargets: optional(
2235
3022
  "OPENGENI_INTEGRATIONS_ALLOW_PRIVATE_NETWORK_TARGETS"
2236
3023
  ),
2237
3024
  integrationsOauthClientsJson: optional("OPENGENI_INTEGRATIONS_OAUTH_CLIENTS_JSON"),
3025
+ gmailWatchTopicName: optional("OPENGENI_GMAIL_WATCH_TOPIC_NAME"),
2238
3026
  slackClientId: optional("OPENGENI_SLACK_CLIENT_ID"),
2239
3027
  slackClientSecret: optional("OPENGENI_SLACK_CLIENT_SECRET"),
2240
3028
  slackSigningSecret: optional("OPENGENI_SLACK_SIGNING_SECRET"),
3029
+ slackAccessMode: optional("OPENGENI_SLACK_ACCESS_MODE"),
2241
3030
  slackBotDisplayName: optional("OPENGENI_SLACK_BOT_DISPLAY_NAME"),
2242
3031
  slackCommand: optional("OPENGENI_SLACK_COMMAND"),
2243
3032
  googleDriveClientId: optional("OPENGENI_GOOGLE_DRIVE_CLIENT_ID"),
@@ -2272,9 +3061,6 @@ function getSettings(source = process.env) {
2272
3061
  goalIdleBackoffMs: optional("OPENGENI_GOAL_IDLE_BACKOFF_MS"),
2273
3062
  goalIdleBackoffMaxMs: optional("OPENGENI_GOAL_IDLE_BACKOFF_MAX_MS"),
2274
3063
  childLifecycleNoticesEnabled: optional("OPENGENI_CHILD_LIFECYCLE_NOTICES_ENABLED"),
2275
- hostMcpAuthoritySourceAdmissionEnabled: optional(
2276
- "OPENGENI_HOST_MCP_AUTHORITY_SOURCE_ADMISSION_ENABLED"
2277
- ),
2278
3064
  slackWorkspaceRoutingEnabled: optional("OPENGENI_SLACK_WORKSPACE_ROUTING_ENABLED"),
2279
3065
  agentMaxModelCallsPerTurn: optional("OPENGENI_AGENT_MAX_MODEL_CALLS_PER_TURN"),
2280
3066
  contextWindowTokens: optional("OPENGENI_CONTEXT_WINDOW_TOKENS"),
@@ -2289,10 +3075,12 @@ function getSettings(source = process.env) {
2289
3075
  authAllowMetrics: optional("OPENGENI_AUTH_ALLOW_METRICS"),
2290
3076
  apiHost: optional("OPENGENI_API_HOST"),
2291
3077
  apiPort: optional("OPENGENI_API_PORT"),
3078
+ apiMetricsPort: optional("OPENGENI_API_METRICS_PORT"),
2292
3079
  workerHttpPort: optional("OPENGENI_WORKER_HTTP_PORT"),
2293
3080
  opengeniMcpInternalUrl: optional("OPENGENI_MCP_INTERNAL_URL"),
2294
3081
  opengeniMcpUrl: optional("OPENGENI_MCP_URL"),
2295
3082
  corsAllowOriginRegex: optional("OPENGENI_CORS_ALLOW_ORIGIN_REGEX"),
3083
+ localAllowedOrigins: optional("OPENGENI_LOCAL_ALLOWED_ORIGINS"),
2296
3084
  openaiProvider: optional("OPENGENI_OPENAI_PROVIDER"),
2297
3085
  openaiApiKey: optional("OPENGENI_OPENAI_API_KEY") ?? optional("OPENAI_API_KEY"),
2298
3086
  openaiBaseUrl: optional("OPENGENI_OPENAI_BASE_URL") ?? optional("OPENAI_BASE_URL"),
@@ -2351,6 +3139,7 @@ function getSettings(source = process.env) {
2351
3139
  modelProvidersJson: optional("OPENGENI_MODEL_PROVIDERS_JSON"),
2352
3140
  codexSubscriptionEnabled: optional("OPENGENI_CODEX_SUBSCRIPTION_ENABLED"),
2353
3141
  supergrokSubscriptionEnabled: optional("OPENGENI_SUPERGROK_SUBSCRIPTION_ENABLED"),
3142
+ claudeSubscriptionEnabled: optional("OPENGENI_CLAUDE_SUBSCRIPTION_ENABLED"),
2354
3143
  supergrokResponseStreamIdleTimeoutMs: optional(
2355
3144
  "OPENGENI_SUPERGROK_RESPONSE_STREAM_IDLE_TIMEOUT_MS"
2356
3145
  ),
@@ -2359,13 +3148,23 @@ function getSettings(source = process.env) {
2359
3148
  lazyToolSearchEnabled: optional("OPENGENI_LAZY_TOOL_SEARCH_ENABLED"),
2360
3149
  codexFleetPolicyShadowEnabled: optional("OPENGENI_CODEX_FLEET_POLICY_SHADOW_ENABLED"),
2361
3150
  codexProductSku: optional("OPENGENI_CODEX_PRODUCT_SKU"),
3151
+ reasoningConfigurationUpdatesEnabled: optional(
3152
+ "OPENGENI_REASONING_CONFIGURATION_UPDATES_ENABLED"
3153
+ ),
2362
3154
  openaiReasoningEffort: optional("OPENGENI_OPENAI_REASONING_EFFORT"),
2363
3155
  openaiAllowedReasoningEfforts: optional("OPENGENI_OPENAI_ALLOWED_REASONING_EFFORTS"),
3156
+ creditsDefaultModel: optional("OPENGENI_CREDITS_DEFAULT_MODEL"),
3157
+ creditsDefaultReasoningEffort: optional("OPENGENI_CREDITS_DEFAULT_REASONING_EFFORT"),
2364
3158
  openaiResponsesTransport: optional("OPENGENI_OPENAI_RESPONSES_TRANSPORT"),
2365
3159
  openaiProviderItemIds: optional("OPENGENI_OPENAI_PROVIDER_ITEM_IDS"),
2366
3160
  openaiReasoningEncryptedContent: optional("OPENGENI_OPENAI_REASONING_ENCRYPTED_CONTENT"),
2367
3161
  openaiMaxRetries: optional("OPENGENI_OPENAI_MAX_RETRIES"),
2368
3162
  webSearchEnabled: optional("OPENGENI_WEB_SEARCH_ENABLED"),
3163
+ jevApiKey: optional("OPENGENI_JEV_API_KEY"),
3164
+ jevBaseUrl: optional("OPENGENI_JEV_BASE_URL"),
3165
+ jevModel: optional("OPENGENI_JEV_MODEL"),
3166
+ jevRequestTimeoutMs: optional("OPENGENI_JEV_REQUEST_TIMEOUT_MS"),
3167
+ codeSearchMode: optional("OPENGENI_CODE_SEARCH_MODE"),
2369
3168
  agentInstructionsTemplate: optional("OPENGENI_AGENT_INSTRUCTIONS_TEMPLATE"),
2370
3169
  azureOpenaiBaseUrl: optional("OPENGENI_AZURE_OPENAI_BASE_URL"),
2371
3170
  azureOpenaiEndpoint: optional("OPENGENI_AZURE_OPENAI_ENDPOINT"),
@@ -2382,6 +3181,7 @@ function getSettings(source = process.env) {
2382
3181
  dockerWorkspaceBaseDir: optional("OPENGENI_DOCKER_WORKSPACE_BASE_DIR"),
2383
3182
  modalAppName: optional("OPENGENI_MODAL_APP_NAME"),
2384
3183
  modalImageRef: optional("OPENGENI_MODAL_IMAGE_REF") ?? DEFAULT_MODAL_IMAGE_REF,
3184
+ modalCommandSupervisionEnabled: optional("OPENGENI_MODAL_COMMAND_SUPERVISION_ENABLED"),
2385
3185
  modalImageId: optional("OPENGENI_MODAL_IMAGE_ID"),
2386
3186
  modalImageRegistrySecret: optional("OPENGENI_MODAL_IMAGE_REGISTRY_SECRET"),
2387
3187
  modalTimeoutSeconds: optional("OPENGENI_MODAL_TIMEOUT_SECONDS"),
@@ -2433,6 +3233,9 @@ function getSettings(source = process.env) {
2433
3233
  blaxelTtl: optional("OPENGENI_BLAXEL_TTL"),
2434
3234
  cloudflareWorkerUrl: optional("OPENGENI_CLOUDFLARE_WORKER_URL"),
2435
3235
  cloudflareApiKey: optional("OPENGENI_CLOUDFLARE_API_KEY"),
3236
+ experimentalBrowserContextPoolEnabled: optional(
3237
+ "OPENGENI_EXPERIMENTAL_BROWSER_CONTEXT_POOL_ENABLED"
3238
+ ),
2436
3239
  browserbaseApiKey: optional("OPENGENI_BROWSERBASE_API_KEY"),
2437
3240
  kernelApiKey: optional("OPENGENI_KERNEL_API_KEY"),
2438
3241
  kernelEndpoint: optional("OPENGENI_KERNEL_ENDPOINT"),
@@ -2481,6 +3284,7 @@ function getSettings(source = process.env) {
2481
3284
  sandboxViewerHolderTtlMs: optional("OPENGENI_SANDBOX_VIEWER_HOLDER_TTL_MS"),
2482
3285
  sandboxInteractionHolderTtlMs: optional("OPENGENI_SANDBOX_INTERACTION_HOLDER_TTL_MS"),
2483
3286
  sandboxIdleGraceMs: optional("OPENGENI_SANDBOX_IDLE_GRACE_MS"),
3287
+ sandboxIdleCommandContainmentMs: optional("OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS"),
2484
3288
  sandboxSnapshotIntervalMs: optional("OPENGENI_SANDBOX_SNAPSHOT_INTERVAL_MS"),
2485
3289
  sandboxSnapshotTimeoutMs: optional("OPENGENI_SANDBOX_SNAPSHOT_TIMEOUT_MS"),
2486
3290
  sandboxDrainSnapshotTimeoutMs: optional("OPENGENI_SANDBOX_DRAIN_SNAPSHOT_TIMEOUT_MS"),
@@ -2494,9 +3298,11 @@ function getSettings(source = process.env) {
2494
3298
  sandboxWarmRateMicrosPerSecondJson: optional(
2495
3299
  "OPENGENI_SANDBOX_WARM_RATE_MICROS_PER_SECOND_JSON"
2496
3300
  ),
3301
+ sandboxWarmBillingMode: optional("OPENGENI_SANDBOX_WARM_BILLING_MODE"),
2497
3302
  sandboxMaxWarmSecondsPerWorkspace: optional("OPENGENI_SANDBOX_MAX_WARM_SECONDS_PER_WORKSPACE"),
2498
3303
  sandboxPreparationProfiles: optional("OPENGENI_SANDBOX_PREPARATION_PROFILES"),
2499
3304
  sandboxEnvAllowlist: optional("OPENGENI_SANDBOX_ENV_ALLOWLIST"),
3305
+ sandboxImageAllowlist: optional("OPENGENI_SANDBOX_IMAGE_ALLOWLIST"),
2500
3306
  objectStorageEndpoint: optional("OPENGENI_OBJECT_STORAGE_ENDPOINT"),
2501
3307
  objectStorageInternalEndpoint: optional("OPENGENI_OBJECT_STORAGE_INTERNAL_ENDPOINT"),
2502
3308
  objectStorageSandboxEndpoint: optional("OPENGENI_OBJECT_STORAGE_SANDBOX_ENDPOINT"),
@@ -2523,6 +3329,13 @@ function getSettings(source = process.env) {
2523
3329
  documentEmbeddingDimensions: optional("OPENGENI_DOCUMENT_EMBEDDING_DIMENSIONS"),
2524
3330
  documentEmbeddingApiKey: optional("OPENGENI_DOCUMENT_EMBEDDING_API_KEY"),
2525
3331
  documentEmbeddingBaseUrl: optional("OPENGENI_DOCUMENT_EMBEDDING_BASE_URL"),
3332
+ documentEmbeddingBillingMode: optional("OPENGENI_DOCUMENT_EMBEDDING_BILLING_MODE"),
3333
+ documentEmbeddingCreditsActivatedAt: optional(
3334
+ "OPENGENI_DOCUMENT_EMBEDDING_CREDITS_ACTIVATED_AT"
3335
+ ),
3336
+ documentEmbeddingRateMicrosPerMillionBytes: optional(
3337
+ "OPENGENI_DOCUMENT_EMBEDDING_RATE_MICROS_PER_MILLION_BYTES"
3338
+ ),
2526
3339
  documentCurationProvider: optional("OPENGENI_DOCUMENT_CURATION_PROVIDER"),
2527
3340
  documentCurationModel: optional("OPENGENI_DOCUMENT_CURATION_MODEL"),
2528
3341
  documentCurationApiKey: optional("OPENGENI_DOCUMENT_CURATION_API_KEY"),
@@ -2557,7 +3370,9 @@ function getSettings(source = process.env) {
2557
3370
  managedAuthGoogleClientSecret: optional("OPENGENI_MANAGED_AUTH_GOOGLE_CLIENT_SECRET"),
2558
3371
  managedAuthGithubClientId: optional("OPENGENI_MANAGED_AUTH_GITHUB_CLIENT_ID"),
2559
3372
  managedAuthGithubClientSecret: optional("OPENGENI_MANAGED_AUTH_GITHUB_CLIENT_SECRET"),
3373
+ allowedUserEmails: source.OPENGENI_ALLOWED_USER_EMAILS === void 0 ? void 0 : source.OPENGENI_ALLOWED_USER_EMAILS.split(",").map((email) => email.trim()),
2560
3374
  managedAuthSessionSetMode: optional("OPENGENI_MANAGED_AUTH_SESSION_SET_MODE"),
3375
+ managedAuthNewSignupsEnabled: optional("OPENGENI_MANAGED_AUTH_NEW_SIGNUPS_ENABLED"),
2561
3376
  organizationUserSetupEmailTokenTransport: optional(
2562
3377
  "OPENGENI_ORGANIZATION_USER_SETUP_EMAIL_TOKEN_TRANSPORT"
2563
3378
  ),
@@ -2576,12 +3391,54 @@ function getSettings(source = process.env) {
2576
3391
  const settings = {
2577
3392
  ...parsed,
2578
3393
  sandboxIdleGraceMs: raw.sandboxIdleGraceMs === void 0 && parsed.sandboxBackend === "modal" ? Math.min(9e5, Math.floor(parsed.modalTimeoutSeconds * 1e3 / 2)) : parsed.sandboxIdleGraceMs,
2579
- sandboxRotationLeadMs: raw.sandboxRotationLeadMs === void 0 && parsed.sandboxBackend === "modal" ? Math.min(36e5, Math.floor(parsed.modalTimeoutSeconds * 1e3 / 2)) : parsed.sandboxRotationLeadMs,
3394
+ sandboxRotationLeadMs: raw.sandboxRotationLeadMs === void 0 && parsed.sandboxBackend === "modal" ? Math.min(
3395
+ 36e5,
3396
+ Math.max(
3397
+ Math.floor(parsed.modalTimeoutSeconds * 1e3 / 2),
3398
+ SANDBOX_DEADLINE_COMMAND_STOP_GRACE_MS + sandboxArchiveCaptureTimeoutMs({
3399
+ sandboxSnapshotTimeoutMs: Math.max(
3400
+ parsed.sandboxSnapshotTimeoutMs,
3401
+ parsed.sandboxDrainSnapshotTimeoutMs ?? parsed.sandboxSnapshotTimeoutMs
3402
+ )
3403
+ }) + 2 * parsed.sandboxLeaseReaperPeriodMs + 1
3404
+ )
3405
+ ) : parsed.sandboxRotationLeadMs,
2580
3406
  mcpServers: ensureBuiltInMcpServers(parsed)
2581
3407
  };
3408
+ if (raw.sandboxIdleCommandContainmentMs === void 0) {
3409
+ const modalIdleCeilingMs = settings.sandboxBackend === "modal" && settings.modalIdleTimeoutSeconds !== void 0 ? settings.modalIdleTimeoutSeconds * 1e3 - settings.sandboxLeaseReaperPeriodMs - sandboxArchiveCaptureTimeoutMs({
3410
+ sandboxSnapshotTimeoutMs: effectiveSandboxDrainSnapshotTimeoutMs(settings)
3411
+ }) - 1 : Number.POSITIVE_INFINITY;
3412
+ const derived = Math.min(
3413
+ 18e5,
3414
+ Math.floor((settings.sandboxIdleGraceMs + settings.sandboxRotationLeadMs) / 2),
3415
+ modalIdleCeilingMs
3416
+ );
3417
+ if (derived > settings.sandboxIdleGraceMs) {
3418
+ settings.sandboxIdleCommandContainmentMs = derived;
3419
+ } else {
3420
+ settings.sandboxIdleCommandContainmentMs = void 0;
3421
+ if (!idleCommandContainmentDisabledWarned) {
3422
+ idleCommandContainmentDisabledWarned = true;
3423
+ console.warn(
3424
+ `[config] idle command containment disabled: no window fits between OPENGENI_SANDBOX_IDLE_GRACE_MS (${settings.sandboxIdleGraceMs}) and the Modal idle timeout / rotation lead ceiling; set OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS to override.`
3425
+ );
3426
+ }
3427
+ }
3428
+ }
3429
+ warnRetiredOrganizationTenancyActivationSwitch(source);
2582
3430
  validateSettings(settings, source);
2583
3431
  return settings;
2584
3432
  }
3433
+ var retiredOrganizationTenancyActivationSwitchWarned = false;
3434
+ function warnRetiredOrganizationTenancyActivationSwitch(source) {
3435
+ if (retiredOrganizationTenancyActivationSwitchWarned) return;
3436
+ if (source.OPENGENI_ORGANIZATION_TENANCY_CANONICAL_ACTIVATION_ENABLED === void 0) return;
3437
+ retiredOrganizationTenancyActivationSwitchWarned = true;
3438
+ console.warn(
3439
+ "[config] OPENGENI_ORGANIZATION_TENANCY_CANONICAL_ACTIVATION_ENABLED is retired and ignored: every organization is session-tenancy activated (migration 0611). Remove it from the deployment."
3440
+ );
3441
+ }
2585
3442
  var LOCAL_FIRST_PARTY_DELEGATION_SECRET = "opengeni-local-first-party-delegation-secret-v1";
2586
3443
  function resolveFirstPartyDelegationSecret(settings) {
2587
3444
  const explicit = settings.delegationSecret?.trim();
@@ -2593,20 +3450,24 @@ function resolveFirstPartyDelegationSecret(settings) {
2593
3450
  return settings.productAccessMode === "local" && (settings.environment === "local" || settings.environment === "test") ? LOCAL_FIRST_PARTY_DELEGATION_SECRET : void 0;
2594
3451
  }
2595
3452
  function resolveFirstPartyMcpToolPolicy(settings) {
2596
- const allowed = settings.allowedFirstPartyMcpTools ?? [...FIRST_PARTY_MCP_TOOL_NAMES];
2597
- const allowedSet = new Set(allowed);
2598
- const defaults = settings.defaultFirstPartyMcpTools ?? [...DEFAULT_FIRST_PARTY_MCP_TOOLS];
3453
+ const allowed2 = currentAgentLearningToolSelection(
3454
+ settings.allowedFirstPartyMcpTools ?? [...FIRST_PARTY_MCP_TOOL_NAMES]
3455
+ ).filter((tool) => !isRetiredNativeAtlassianTool(tool));
3456
+ const allowedSet = new Set(allowed2);
3457
+ const defaults = currentAgentLearningToolSelection(
3458
+ settings.defaultFirstPartyMcpTools ?? [...DEFAULT_FIRST_PARTY_MCP_TOOLS]
3459
+ );
2599
3460
  return {
2600
3461
  default: defaults.filter((tool) => allowedSet.has(tool)),
2601
- allowed: [...allowed]
3462
+ allowed: [...allowed2]
2602
3463
  };
2603
3464
  }
2604
3465
  function allowedFirstPartyMcpToolsForSession(settings, selected) {
2605
3466
  const policy = resolveFirstPartyMcpToolPolicy(settings);
2606
- const allowed = new Set(policy.allowed);
2607
- const tools = new Set(selected ?? policy.default);
3467
+ const allowed2 = new Set(policy.allowed);
3468
+ const tools = new Set(currentAgentLearningToolSelection(selected ?? policy.default));
2608
3469
  if (tools.has("goal_pause")) tools.add("goal_resume");
2609
- return [...tools].filter((tool) => allowed.has(tool));
3470
+ return [...tools].filter((tool) => allowed2.has(tool));
2610
3471
  }
2611
3472
  function effectiveModalIdleTimeoutSeconds(settings) {
2612
3473
  return settings.modalIdleTimeoutSeconds ?? settings.modalTimeoutSeconds;
@@ -3087,7 +3948,9 @@ function openRouterRegistryProvider(settings, input) {
3087
3948
  }
3088
3949
  function configuredRegistryProviders(settings) {
3089
3950
  const providers = parseModelProvidersJson(settings.modelProvidersJson);
3090
- const injected = [...providers];
3951
+ const injected = providers.filter(
3952
+ (provider) => settings.claudeSubscriptionEnabled || !(provider.api === "anthropic-messages" && provider.anthropic?.auth === "oauth")
3953
+ );
3091
3954
  if (settings.vercelAiGatewayApiKey && configuredGatewayCatalogModels(settings).length > 0) {
3092
3955
  injected.push(
3093
3956
  gatewayRegistryProvider(settings, {
@@ -3103,6 +3966,37 @@ function configuredRegistryProviders(settings) {
3103
3966
  if (openrouter) injected.push(openrouter);
3104
3967
  return injected;
3105
3968
  }
3969
+ function withDirectModelProviders(settings, connections) {
3970
+ const providers = parseModelProvidersJson(settings.modelProvidersJson).filter(
3971
+ (provider) => !provider.id.startsWith("workspace-openai-") && !provider.id.startsWith("workspace-azure-openai-")
3972
+ );
3973
+ for (const connection of connections) {
3974
+ const spec = directModelConnectionSpec(connection);
3975
+ if (!spec) continue;
3976
+ providers.push({
3977
+ kind: spec.provider === "openai" ? "direct-openai-workspace" : "direct-azure-workspace",
3978
+ id: spec.providerId,
3979
+ label: spec.provider === "openai" ? "Your OpenAI" : "Your Azure OpenAI",
3980
+ api: "responses",
3981
+ wireProfile: spec.provider === "openai" ? "openai" : "azure-openai",
3982
+ baseUrl: spec.baseUrl,
3983
+ ...connection.apiKey ? { apiKey: connection.apiKey } : {},
3984
+ models: [
3985
+ {
3986
+ id: spec.modelId,
3987
+ upstreamModelId: spec.model,
3988
+ label: spec.model,
3989
+ capabilities: legacyModelCapabilities(settings, {
3990
+ reasoningEffort: false,
3991
+ hostedWebSearch: false
3992
+ }),
3993
+ toolOutputTruncationTokens: settings.modelToolOutputTruncationTokens
3994
+ }
3995
+ ]
3996
+ });
3997
+ }
3998
+ return { ...settings, modelProvidersJson: JSON.stringify(providers) };
3999
+ }
3106
4000
  function withWorkspaceGatewayCatalogProvider(settings, customModels = []) {
3107
4001
  const providers = parseModelProvidersJson(settings.modelProvidersJson);
3108
4002
  const withoutWorkspace = providers.filter(
@@ -3215,8 +4109,8 @@ function productLabelForModelId(modelId) {
3215
4109
  function productShortLabelForModelId(modelId) {
3216
4110
  const slug = modelId.startsWith(CODEX_MODEL_ID_PREFIX) ? modelId.slice(CODEX_MODEL_ID_PREFIX.length) : modelId;
3217
4111
  switch (slug) {
3218
- case "grok-4.6":
3219
- return "4.6";
4112
+ case "grok-4.7":
4113
+ return "4.7";
3220
4114
  case "gpt-5.6-sol":
3221
4115
  return "5.6 Sol";
3222
4116
  case "gpt-5.6-terra":
@@ -3225,6 +4119,10 @@ function productShortLabelForModelId(modelId) {
3225
4119
  return "5.6 Luna";
3226
4120
  case "gpt-6-astra":
3227
4121
  return "6 Astra";
4122
+ case "gpt-6-sol":
4123
+ return "6 Sol";
4124
+ case "gpt-6-luna":
4125
+ return "6 Luna";
3228
4126
  default:
3229
4127
  return null;
3230
4128
  }
@@ -3244,7 +4142,7 @@ function builtinContextLimitsForModel(settings, modelId) {
3244
4142
  return { contextWindowTokens: settings.contextWindowTokens };
3245
4143
  }
3246
4144
  function builtinLatencyModesForModel(modelId) {
3247
- if (isBuiltinGpt56ModelId(modelId) || modelId.startsWith("codex/gpt-5.6-") || modelId === "codex/gpt-6-astra") {
4145
+ if (isBuiltinGpt56ModelId(modelId) || /^gpt-6(?:\.\d+)?-/u.test(modelId) || modelId.startsWith("codex/gpt-5.6-") || /^codex\/gpt-6(?:\.\d+)?-/u.test(modelId)) {
3248
4146
  return [
3249
4147
  { id: "standard", upstream: "supported", runnable: true },
3250
4148
  {
@@ -3257,12 +4155,26 @@ function builtinLatencyModesForModel(modelId) {
3257
4155
  }
3258
4156
  return [{ id: "standard", upstream: "unknown", runnable: true }];
3259
4157
  }
4158
+ function isReviewedGptVisionModel(slug) {
4159
+ return slug.startsWith("gpt-5.6-") || /^gpt-6(?:\.\d+)?-/u.test(slug);
4160
+ }
4161
+ function reviewedCodexCatalogCapabilities(slug, capabilities) {
4162
+ if (!/^gpt-6(?:\.\d+)?-/u.test(slug)) return capabilities;
4163
+ const fast = builtinLatencyModesForModel(`${CODEX_MODEL_ID_PREFIX}${slug}`).find(
4164
+ (mode) => mode.id === "fast"
4165
+ );
4166
+ return normalizeCapabilities({
4167
+ ...capabilities,
4168
+ inputModalities: [.../* @__PURE__ */ new Set([...capabilities.inputModalities, "image"])],
4169
+ latencyModes: fast && !capabilities.latencyModes.some((mode) => mode.id === "fast") ? [...capabilities.latencyModes, fast] : capabilities.latencyModes
4170
+ });
4171
+ }
3260
4172
  function builtinPromptCachingForModel(modelId) {
3261
4173
  const slug = modelId.startsWith(CODEX_MODEL_ID_PREFIX) ? modelId.slice(CODEX_MODEL_ID_PREFIX.length) : modelId;
3262
- return slug.startsWith("gpt-5.6-") ? { upstream: "supported", runnable: true, mode: "implicit" } : void 0;
4174
+ return slug.startsWith("gpt-5.6-") || slug.startsWith("gpt-6-") ? { upstream: "supported", runnable: true, mode: "implicit" } : void 0;
3263
4175
  }
3264
4176
  function builtinHostedImageGenerationForModel(settings, modelId) {
3265
- return settings.openaiProvider === "openai" && isDirectOpenAiApiBaseUrl(settings.openaiBaseUrl) && isBuiltinGpt56ModelId(modelId);
4177
+ return settings.openaiProvider === "openai" && isDirectOpenAiApiBaseUrl(settings.openaiBaseUrl) && (isBuiltinGpt56ModelId(modelId) || modelId.startsWith("gpt-6-"));
3266
4178
  }
3267
4179
  function isDirectOpenAiApiBaseUrl(baseUrl) {
3268
4180
  if (baseUrl === void 0) return true;
@@ -3301,11 +4213,21 @@ function runnableLatencyModesForModel(settings, modelId) {
3301
4213
  function assertLatencyModeRunnable(settings, modelId, latencyMode) {
3302
4214
  const runnable = runnableLatencyModesForModel(settings, modelId);
3303
4215
  if (!runnable.includes(latencyMode)) {
3304
- throw new Error(
3305
- `latency mode ${latencyMode} is not runnable for model ${modelId} (allowed: ${runnable.join(", ")})`
3306
- );
4216
+ throw new UnsupportedLatencyModeError(modelId, latencyMode, runnable);
3307
4217
  }
3308
4218
  }
4219
+ var UnsupportedLatencyModeError = class extends Error {
4220
+ constructor(modelId, latencyMode, allowedLatencyModes) {
4221
+ super(
4222
+ `latency mode ${latencyMode} is not runnable for model ${modelId} (allowed: ${allowedLatencyModes.join(", ")})`
4223
+ );
4224
+ this.modelId = modelId;
4225
+ this.latencyMode = latencyMode;
4226
+ this.allowedLatencyModes = allowedLatencyModes;
4227
+ this.name = "UnsupportedLatencyModeError";
4228
+ }
4229
+ code = "UNSUPPORTED_LATENCY_MODE";
4230
+ };
3309
4231
  function registryCredentialSource(provider) {
3310
4232
  switch (provider.kind) {
3311
4233
  case "anonymous":
@@ -3314,10 +4236,17 @@ function registryCredentialSource(provider) {
3314
4236
  return { kind: "connected_subscription", provider: "codex" };
3315
4237
  case "xai-subscription":
3316
4238
  return { kind: "connected_subscription", provider: "xai" };
4239
+ case "claude-subscription-workspace":
4240
+ case "claude-subscription-organization":
4241
+ return { kind: "connected_subscription", provider: "claude" };
3317
4242
  case "vercel-gateway-workspace":
4243
+ case "direct-openai-workspace":
4244
+ case "direct-azure-workspace":
3318
4245
  case "openrouter-workspace":
4246
+ case "anthropic-workspace":
3319
4247
  return { kind: "workspace_connection", mechanism: "api_key" };
3320
4248
  case "vercel-gateway-organization":
4249
+ case "anthropic-organization":
3321
4250
  case "openrouter-organization":
3322
4251
  return { kind: "organization_connection", mechanism: "api_key" };
3323
4252
  case "api-key":
@@ -3337,11 +4266,17 @@ function registryBilling(provider) {
3337
4266
  return { upstreamPayer: "deployment", metering: "external" };
3338
4267
  case "codex-subscription":
3339
4268
  case "xai-subscription":
4269
+ case "claude-subscription-workspace":
4270
+ case "claude-subscription-organization":
3340
4271
  return { upstreamPayer: "connected_subscription", metering: "external" };
3341
4272
  case "vercel-gateway-workspace":
4273
+ case "direct-openai-workspace":
4274
+ case "direct-azure-workspace":
3342
4275
  case "openrouter-workspace":
4276
+ case "anthropic-workspace":
3343
4277
  return { upstreamPayer: "workspace", metering: "external" };
3344
4278
  case "vercel-gateway-organization":
4279
+ case "anthropic-organization":
3345
4280
  case "openrouter-organization":
3346
4281
  return { upstreamPayer: "organization", metering: "external" };
3347
4282
  case "api-key":
@@ -3372,6 +4307,9 @@ function staticRequestMetadataForDigest(provider) {
3372
4307
  const publicHeaders = new Set(provider.publicDefaultHeaderNames ?? []);
3373
4308
  const publicQuery = new Set(provider.publicDefaultQueryNames ?? []);
3374
4309
  return {
4310
+ ...provider.anthropic ? {
4311
+ anthropic: (({ identity: _identity, credentialBinding: _binding, ...options }) => options)(provider.anthropic)
4312
+ } : {},
3375
4313
  headers: Object.entries(provider.defaultHeaders ?? {}).sort(([left], [right]) => left.localeCompare(right)).map(
3376
4314
  ([name, value]) => publicHeaders.has(name) ? { name, classification: "public", value } : { name, classification: "secret" }
3377
4315
  ),
@@ -3413,7 +4351,8 @@ function definitionVersionFor(model, provider, options = {}) {
3413
4351
  ...includeWireProfile ? { wireProfile: provider.wireProfile } : {},
3414
4352
  baseUrl: provider.baseUrl ?? null,
3415
4353
  defaultHeaders: requestMetadata.headers,
3416
- defaultQuery: requestMetadata.query
4354
+ defaultQuery: requestMetadata.query,
4355
+ ...requestMetadata.anthropic ? { anthropic: requestMetadata.anthropic } : {}
3417
4356
  },
3418
4357
  credentialSource: model.credentialSource,
3419
4358
  billing: model.billing,
@@ -3435,9 +4374,42 @@ function legacyImplicitOpenAiDefinitionVersionFor(model, provider) {
3435
4374
  includeWireProfile: false
3436
4375
  });
3437
4376
  }
3438
- function builtinProviderId(settings) {
3439
- return settings.openaiProvider === "azure" ? "azure" : "openai";
3440
- }
4377
+ function legacyCodexAstraImplicitCachingDefinitionVersionFor(model, provider) {
4378
+ if (model.id !== "codex/gpt-6-astra" || model.upstreamModelId !== "gpt-6-astra" || provider.id !== CODEX_PROVIDER_ID || provider.kind !== "codex-subscription" || provider.api !== "responses" || provider.wireProfile !== "openai" || canonicalJson(model.capabilities.promptCaching) !== canonicalJson({ upstream: "supported", runnable: true, mode: "implicit" })) {
4379
+ return null;
4380
+ }
4381
+ const { definitionVersion: _definitionVersion, ...modelWithoutVersion } = model;
4382
+ const { promptCaching: _promptCaching, ...capabilities } = model.capabilities;
4383
+ return definitionVersionFor({ ...modelWithoutVersion, capabilities }, provider);
4384
+ }
4385
+ function matchesAdditiveCapabilityDefinitionVersion(model, provider, policy) {
4386
+ const { definitionVersion: _definitionVersion, ...modelWithoutVersion } = model;
4387
+ const { latencyModes, inputModalities } = model.capabilities;
4388
+ for (let latencyMask = 1; latencyMask < 1 << latencyModes.length; latencyMask += 1) {
4389
+ const retainedModes = latencyModes.filter((_mode, index) => latencyMask & 1 << index);
4390
+ if (!retainedModes.some((mode) => mode.id === policy.latencyMode && mode.runnable)) {
4391
+ continue;
4392
+ }
4393
+ for (let inputMask = 1; inputMask < 1 << inputModalities.length; inputMask += 1) {
4394
+ const retainedInputs = inputModalities.filter((_modality, index) => inputMask & 1 << index);
4395
+ if (retainedModes.length === latencyModes.length && retainedInputs.length === inputModalities.length) {
4396
+ continue;
4397
+ }
4398
+ const capabilities = {
4399
+ ...model.capabilities,
4400
+ latencyModes: retainedModes,
4401
+ inputModalities: retainedInputs
4402
+ };
4403
+ if (policy.definitionVersion === definitionVersionFor({ ...modelWithoutVersion, capabilities }, provider)) {
4404
+ return true;
4405
+ }
4406
+ }
4407
+ }
4408
+ return false;
4409
+ }
4410
+ function builtinProviderId(settings) {
4411
+ return settings.openaiProvider === "azure" ? "azure" : "openai";
4412
+ }
3441
4413
  function builtinProviderLabel(settings) {
3442
4414
  return settings.openaiProvider === "azure" ? "Azure OpenAI" : "OpenAI";
3443
4415
  }
@@ -3456,10 +4428,10 @@ function configuredProviders(settings, source = process.env) {
3456
4428
  if (settings.openaiProvider === "azure") {
3457
4429
  const baseUrl = settings.azureOpenaiBaseUrl ?? settings.azureOpenaiEndpoint;
3458
4430
  builtin.baseUrl = baseUrl ? normalizeRegistryBaseUrl(baseUrl, builtin.id) : void 0;
3459
- builtin.apiKey = settings.azureOpenaiApiKey ?? settings.azureOpenaiAdToken;
4431
+ builtin.apiKey = usableDeploymentSecret(settings.azureOpenaiApiKey) ?? settings.azureOpenaiAdToken;
3460
4432
  } else {
3461
4433
  builtin.baseUrl = settings.openaiBaseUrl ? normalizeRegistryBaseUrl(settings.openaiBaseUrl, builtin.id) : void 0;
3462
- builtin.apiKey = settings.openaiApiKey;
4434
+ builtin.apiKey = usableDeploymentSecret(settings.openaiApiKey);
3463
4435
  }
3464
4436
  const registry = configuredRegistryProviders(settings).map(
3465
4437
  (provider) => ({
@@ -3467,6 +4439,7 @@ function configuredProviders(settings, source = process.env) {
3467
4439
  label: provider.label ?? provider.id,
3468
4440
  kind: provider.kind,
3469
4441
  api: provider.api,
4442
+ anthropic: provider.anthropic,
3470
4443
  wireProfile: provider.wireProfile,
3471
4444
  builtin: false,
3472
4445
  baseUrl: provider.baseUrl,
@@ -3481,11 +4454,17 @@ function configuredProviders(settings, source = process.env) {
3481
4454
  );
3482
4455
  return [builtin, ...registry];
3483
4456
  }
4457
+ function codexUpstreamModelSlugs(settings) {
4458
+ if (settings.resolvedCodexModelsJson === void 0) return [...CODEX_FALLBACK_MODEL_SLUGS];
4459
+ return z2.array(CodexCatalogModelSchema).parse(JSON.parse(settings.resolvedCodexModelsJson)).filter((model) => !model.retired).map((model) => model.upstreamModelId);
4460
+ }
3484
4461
  function withCodexCatalogProvider(settings) {
3485
4462
  const providers = parseModelProvidersJson(settings.modelProvidersJson);
3486
4463
  if (providers.some((provider2) => provider2.id === CODEX_PROVIDER_ID)) {
3487
4464
  return settings;
3488
4465
  }
4466
+ const catalogModels = settings.resolvedCodexModelsJson === void 0 ? void 0 : z2.array(CodexCatalogModelSchema).parse(JSON.parse(settings.resolvedCodexModelsJson)).filter((model) => !model.retired).map(({ retired: _retired, ...model }) => model);
4467
+ if (catalogModels?.length === 0) return settings;
3489
4468
  const provider = {
3490
4469
  kind: "codex-subscription",
3491
4470
  id: CODEX_PROVIDER_ID,
@@ -3493,12 +4472,12 @@ function withCodexCatalogProvider(settings) {
3493
4472
  api: "responses",
3494
4473
  wireProfile: "openai",
3495
4474
  baseUrl: CODEX_PROVIDER_BASE_URL,
3496
- models: CODEX_FALLBACK_MODEL_SLUGS.map((slug) => {
4475
+ models: catalogModels ?? CODEX_FALLBACK_MODEL_SLUGS.map((slug) => {
3497
4476
  const capabilities = {
3498
4477
  ...legacyModelCapabilities(settings, {
3499
4478
  reasoningEffort: true,
3500
4479
  hostedWebSearch: true,
3501
- vision: slug.startsWith("gpt-5.6-") || slug === "gpt-6-astra"
4480
+ vision: isReviewedGptVisionModel(slug)
3502
4481
  }),
3503
4482
  ...builtinPromptCachingForModel(`${CODEX_MODEL_ID_PREFIX}${slug}`) ? {
3504
4483
  promptCaching: builtinPromptCachingForModel(`${CODEX_MODEL_ID_PREFIX}${slug}`)
@@ -3681,7 +4660,7 @@ function configuredModels(settings, source = process.env) {
3681
4660
  reasoningEffort: true,
3682
4661
  hostedWebSearch: settings.webSearchEnabled,
3683
4662
  hostedImageGeneration: builtinHostedImageGenerationForModel(settings, id),
3684
- vision: id.startsWith("gpt-5.6-")
4663
+ vision: isReviewedGptVisionModel(id)
3685
4664
  }),
3686
4665
  ...builtinPromptCachingForModel(id) ? { promptCaching: builtinPromptCachingForModel(id) } : {},
3687
4666
  latencyModes: builtinLatencyModesForModel(id)
@@ -3760,7 +4739,57 @@ function canonicalizeConfiguredModelId(settings, modelId) {
3760
4739
  return models.find((model) => model.aliases.includes(modelId))?.id ?? modelId;
3761
4740
  }
3762
4741
  function configuredAllowedModels(settings) {
3763
- return configuredModels(settings).map((model) => model.id);
4742
+ return configuredModels(settings).filter((model) => isModelAvailableForNewSelection(settings, model.id)).map((model) => model.id);
4743
+ }
4744
+ function isModelAvailableForNewSelection(settings, modelId) {
4745
+ if (!modelId.startsWith(CODEX_MODEL_ID_PREFIX) || settings.resolvedCodexModelsJson === void 0)
4746
+ return true;
4747
+ const models = z2.array(CodexCatalogModelSchema).parse(JSON.parse(settings.resolvedCodexModelsJson));
4748
+ return models.some((model) => model.id === modelId && !model.retired);
4749
+ }
4750
+ function settingsForAcceptedSubscriptionTurn(settings, policy, expected) {
4751
+ const parsed = TurnExecutionPolicyV1.parse(policy);
4752
+ if (!parsed.productModelId.startsWith(CODEX_MODEL_ID_PREFIX)) return settings;
4753
+ if (!settings.codexSubscriptionEnabled) throw new Error("Codex subscription is disabled");
4754
+ const models = settings.resolvedCodexModelsJson === void 0 ? [] : z2.array(CodexCatalogModelSchema).parse(JSON.parse(settings.resolvedCodexModelsJson));
4755
+ const retained = models.find((model2) => model2.id === parsed.productModelId && model2.retired);
4756
+ if (!retained) {
4757
+ assertTurnExecutionPolicyMatchesConfigV1(settings, parsed, expected);
4758
+ return settings;
4759
+ }
4760
+ const { retired: _retired, ...model } = retained;
4761
+ const withoutCodex = {
4762
+ ...settings,
4763
+ modelProvidersJson: JSON.stringify(
4764
+ parseModelProvidersJson(settings.modelProvidersJson).filter(
4765
+ (provider) => provider.id !== CODEX_PROVIDER_ID
4766
+ )
4767
+ )
4768
+ };
4769
+ const providerSettings = withCodexCatalogProvider({
4770
+ ...withoutCodex,
4771
+ resolvedCodexModelsJson: JSON.stringify([model])
4772
+ });
4773
+ const retainedProvider = parseModelProvidersJson(providerSettings.modelProvidersJson).find(
4774
+ (provider) => provider.id === CODEX_PROVIDER_ID
4775
+ );
4776
+ const providers = parseModelProvidersJson(settings.modelProvidersJson);
4777
+ const current = providers.find((provider) => provider.id === CODEX_PROVIDER_ID);
4778
+ const restored = {
4779
+ ...settings,
4780
+ modelProvidersJson: JSON.stringify([
4781
+ ...providers.filter((provider) => provider.id !== CODEX_PROVIDER_ID),
4782
+ {
4783
+ ...retainedProvider,
4784
+ models: [
4785
+ ...(current?.models ?? []).filter((candidate) => candidate.id !== model.id),
4786
+ model
4787
+ ]
4788
+ }
4789
+ ])
4790
+ };
4791
+ assertTurnExecutionPolicyMatchesConfigV1(restored, parsed, expected);
4792
+ return restored;
3764
4793
  }
3765
4794
  function resolveModelProvider(settings, modelId) {
3766
4795
  const canonicalModelId = canonicalizeConfiguredModelId(settings, modelId);
@@ -3810,12 +4839,25 @@ function resolveModelProviderForTurn(settings, modelId) {
3810
4839
  canonicalizeConfiguredModelId(catalogSettings, modelId)
3811
4840
  );
3812
4841
  }
4842
+ var TurnExecutionPolicyModelUnavailableError = class extends Error {
4843
+ constructor(message) {
4844
+ super(message);
4845
+ this.name = "TurnExecutionPolicyModelUnavailableError";
4846
+ }
4847
+ };
3813
4848
  function resolveTurnExecutionPolicyV1(settings, input) {
3814
4849
  const catalogSettings = settingsForTurnExecutionPolicy(settings, input.modelId);
3815
4850
  const productModelId = canonicalizeConfiguredModelId(catalogSettings, input.modelId);
4851
+ if (!isModelAvailableForNewSelection(catalogSettings, productModelId)) {
4852
+ throw new TurnExecutionPolicyModelUnavailableError(
4853
+ "Turn execution policy model is retired from new selection"
4854
+ );
4855
+ }
3816
4856
  const resolved = resolveModelProvider(catalogSettings, productModelId);
3817
4857
  if (!resolved) {
3818
- throw new Error("Turn execution policy model is not present in the configured catalog");
4858
+ throw new TurnExecutionPolicyModelUnavailableError(
4859
+ "Turn execution policy model is not present in the configured catalog"
4860
+ );
3819
4861
  }
3820
4862
  if (input.requestedModelId !== null && canonicalizeConfiguredModelId(catalogSettings, input.requestedModelId) !== productModelId) {
3821
4863
  throw new Error("Turn execution policy requested model does not canonicalize to its product");
@@ -3840,6 +4882,13 @@ function resolveTurnExecutionPolicyV1(settings, input) {
3840
4882
  definitionVersion: resolved.model.definitionVersion
3841
4883
  });
3842
4884
  }
4885
+ var TurnExecutionPolicyDefinitionMismatchError = class extends Error {
4886
+ code = "turn_execution_policy_definition_mismatch";
4887
+ constructor() {
4888
+ super("Turn execution policy does not match the current provider definition");
4889
+ this.name = "TurnExecutionPolicyDefinitionMismatchError";
4890
+ }
4891
+ };
3843
4892
  function assertTurnExecutionPolicyMatchesConfigV1(settings, policy, expected) {
3844
4893
  const parsed = TurnExecutionPolicyV1.parse(policy);
3845
4894
  const catalogSettings = settingsForTurnExecutionPolicy(settings, parsed.productModelId);
@@ -3858,15 +4907,33 @@ function assertTurnExecutionPolicyMatchesConfigV1(settings, policy, expected) {
3858
4907
  if (!resolved) {
3859
4908
  throw new Error("Turn execution policy model is no longer configured");
3860
4909
  }
4910
+ const legacyClaudeScope = resolved.provider.kind === "claude-subscription-workspace" ? "workspace" : resolved.provider.kind === "claude-subscription-organization" ? "organization" : null;
4911
+ const legacyClaudeAccountingMatches = legacyClaudeScope !== null && parsed.providerId === resolved.provider.id && parsed.upstreamModelId === resolved.model.upstreamModelId && parsed.wireApi === resolved.model.api && canonicalJson(parsed.credentialSource) === canonicalJson({
4912
+ kind: `${legacyClaudeScope}_connection`,
4913
+ mechanism: "api_key"
4914
+ }) && canonicalJson(parsed.billing) === canonicalJson({
4915
+ upstreamPayer: legacyClaudeScope,
4916
+ metering: "external"
4917
+ }) && parsed.definitionVersion === definitionVersionFor(
4918
+ {
4919
+ ...resolved.model,
4920
+ credentialSource: parsed.credentialSource,
4921
+ billing: parsed.billing
4922
+ },
4923
+ resolved.provider
4924
+ );
3861
4925
  const legacyImplicitOpenAiDefinitionVersion = legacyImplicitOpenAiDefinitionVersionFor(
3862
4926
  resolved.model,
3863
4927
  resolved.provider
3864
4928
  );
3865
- const definitionVersionMatches = parsed.definitionVersion === resolved.model.definitionVersion || parsed.definitionVersion === legacyImplicitOpenAiDefinitionVersion;
3866
- const mismatched = parsed.providerId !== resolved.provider.id || parsed.upstreamModelId !== resolved.model.upstreamModelId || parsed.wireApi !== resolved.model.api || !definitionVersionMatches || canonicalJson(parsed.credentialSource) !== canonicalJson(resolved.model.credentialSource) || canonicalJson(parsed.billing) !== canonicalJson(resolved.model.billing);
3867
- if (mismatched) {
4929
+ const definitionVersionMatches = parsed.definitionVersion === resolved.model.definitionVersion || parsed.definitionVersion === legacyImplicitOpenAiDefinitionVersion || parsed.definitionVersion === legacyCodexAstraImplicitCachingDefinitionVersionFor(resolved.model, resolved.provider) || matchesAdditiveCapabilityDefinitionVersion(resolved.model, resolved.provider, parsed);
4930
+ const identityMismatched = parsed.providerId !== resolved.provider.id || parsed.upstreamModelId !== resolved.model.upstreamModelId || parsed.wireApi !== resolved.model.api || canonicalJson(parsed.credentialSource) !== canonicalJson(resolved.model.credentialSource) || canonicalJson(parsed.billing) !== canonicalJson(resolved.model.billing);
4931
+ if (identityMismatched && !legacyClaudeAccountingMatches) {
3868
4932
  throw new Error("Turn execution policy does not match the current provider definition");
3869
4933
  }
4934
+ if (!definitionVersionMatches && !legacyClaudeAccountingMatches) {
4935
+ throw new TurnExecutionPolicyDefinitionMismatchError();
4936
+ }
3870
4937
  return { policy: parsed, provider: resolved.provider, model: resolved.model };
3871
4938
  }
3872
4939
  function configuredModelPricingSchedules(settings) {
@@ -3896,6 +4963,103 @@ function configuredModelPricingSchedules(settings) {
3896
4963
  ...configured
3897
4964
  };
3898
4965
  }
4966
+ function configuredModelListPricingSchedules(settings) {
4967
+ const prices = configuredModelPricingSchedules(settings);
4968
+ const configured = Object.fromEntries(
4969
+ Object.entries(parseModelPricingJson(settings.modelPricingJson)).map(([model, pricing]) => [
4970
+ model,
4971
+ normalizeModelPricingSchedule(pricing)
4972
+ ])
4973
+ );
4974
+ if (settings.openaiProvider === "openai" && isDirectOpenAiApiBaseUrl(settings.openaiBaseUrl ?? "https://api.openai.com/v1")) {
4975
+ for (const model of configuredModels(settings)) {
4976
+ if (model.providerId !== builtinProviderId(settings) || !model.upstreamModelId.startsWith("gpt-") || prices[model.id] !== void 0)
4977
+ continue;
4978
+ const reviewed = reviewedModelListPricing[model.upstreamModelId];
4979
+ if (reviewed) prices[model.id] = reviewed;
4980
+ }
4981
+ }
4982
+ for (const provider of configuredRegistryProviders(settings)) {
4983
+ for (const model of provider.models) {
4984
+ if (prices[model.id] !== void 0) continue;
4985
+ const reviewed = reviewedProviderModelPricing(settings, provider, model, configured);
4986
+ if (reviewed) prices[model.id] = reviewed;
4987
+ }
4988
+ }
4989
+ return prices;
4990
+ }
4991
+ function reviewedProviderModelPricing(settings, provider, model, configured) {
4992
+ const upstream = model.upstreamModelId ?? model.id;
4993
+ let priceId;
4994
+ let nativeClaude = false;
4995
+ switch (provider.kind) {
4996
+ case "anonymous":
4997
+ if (provider.baseUrl.replace(/\/$/u, "") === OPENROUTER_BASE_URL && upstream === DEFAULT_OPENROUTER_MODEL_ID.slice(OPENROUTER_MODEL_ID_PREFIX.length)) {
4998
+ priceId = DEFAULT_OPENROUTER_MODEL_ID;
4999
+ }
5000
+ break;
5001
+ case "codex-subscription":
5002
+ if (upstream.startsWith("gpt-")) priceId = upstream;
5003
+ break;
5004
+ case "xai-subscription":
5005
+ if (upstream.startsWith("grok-")) priceId = upstream;
5006
+ break;
5007
+ case "anthropic-workspace":
5008
+ case "anthropic-organization":
5009
+ case "claude-subscription-workspace":
5010
+ case "claude-subscription-organization":
5011
+ if (claudeNativeModelProfile(upstream)) {
5012
+ priceId = upstream;
5013
+ nativeClaude = true;
5014
+ }
5015
+ break;
5016
+ case "direct-openai-workspace":
5017
+ if (isDirectOpenAiApiBaseUrl(provider.baseUrl) && upstream.startsWith("gpt-")) {
5018
+ priceId = upstream;
5019
+ }
5020
+ break;
5021
+ case "vercel-gateway-managed":
5022
+ case "vercel-gateway-workspace":
5023
+ case "vercel-gateway-organization":
5024
+ priceId = configuredGatewayCatalogModels(settings).find(
5025
+ (candidate) => candidate.upstreamModelId === upstream && Object.values(OPENGENI_GATEWAY_MODELS).some(
5026
+ (reviewed) => reviewed.upstreamModelId === candidate.upstreamModelId
5027
+ )
5028
+ )?.productId;
5029
+ break;
5030
+ case "openrouter-workspace":
5031
+ case "openrouter-organization":
5032
+ case "openrouter-managed":
5033
+ if (upstream === DEFAULT_OPENROUTER_MODEL_ID.slice(OPENROUTER_MODEL_ID_PREFIX.length)) {
5034
+ priceId = DEFAULT_OPENROUTER_MODEL_ID;
5035
+ }
5036
+ break;
5037
+ case "api-key":
5038
+ if (isDirectOpenAiApiBaseUrl(provider.baseUrl) && upstream.startsWith("gpt-")) {
5039
+ priceId = upstream;
5040
+ } else if (provider.baseUrl.replace(/\/$/u, "") === "https://api.x.ai/v1") {
5041
+ if (upstream.startsWith("grok-")) priceId = upstream;
5042
+ } else if (provider.api === "anthropic-messages" && provider.baseUrl.replace(/\/$/u, "") === "https://api.anthropic.com/v1" && claudeNativeModelProfile(upstream)) {
5043
+ priceId = upstream;
5044
+ nativeClaude = true;
5045
+ }
5046
+ break;
5047
+ default:
5048
+ return void 0;
5049
+ }
5050
+ if (!priceId) return void 0;
5051
+ if (configured[priceId]) return configured[priceId];
5052
+ const schedule = reviewedModelListPricing[priceId] ?? defaultModelPricing[priceId] ?? (priceId === DEFAULT_OPENROUTER_MODEL_ID ? reviewedFreeOpenRouterListPricing : void 0);
5053
+ if (!schedule) return void 0;
5054
+ if (!nativeClaude || provider.anthropic?.cacheTtl !== "1h") return schedule;
5055
+ return {
5056
+ ...schedule,
5057
+ default: {
5058
+ ...schedule.default,
5059
+ cacheWriteMicrosPerMillionTokens: schedule.default.inputMicrosPerMillionTokens * 2
5060
+ }
5061
+ };
5062
+ }
3899
5063
  function configuredModelPricing(settings) {
3900
5064
  return Object.fromEntries(
3901
5065
  Object.entries(configuredModelPricingSchedules(settings)).map(([model, schedule]) => [
@@ -3961,6 +5125,216 @@ function calculateModelUsageCostMicros(settings, model, usage, options) {
3961
5125
  }
3962
5126
  function calculateModelUsageCostBreakdown(settings, model, usage, options) {
3963
5127
  const schedule = configuredModelPricingSchedules(settings)[model];
5128
+ return calculateUsageCostBreakdown(settings, model, usage, schedule, options);
5129
+ }
5130
+ function calculateModelListUsageCostBreakdown(settings, model, usage, options) {
5131
+ const schedule = configuredModelListPricingSchedules(settings)[model];
5132
+ return calculateUsageCostBreakdown(settings, model, usage, schedule, options);
5133
+ }
5134
+ function calculateModelListUsageCostSnapshot(settings, model, usage, options) {
5135
+ const schedule = configuredModelListPricingSchedules(settings)[model];
5136
+ const fallback = calculateUsageCostBreakdown(settings, model, usage, schedule, options);
5137
+ const unknown = { ...fallback, listByClassMicros: null, listByClassApprox: false };
5138
+ if (!schedule || !options?.priceContextKnown) return unknown;
5139
+ const entries = usage.requestUsageEntries?.length ? usage.requestUsageEntries : [usage];
5140
+ const classes = {
5141
+ uncachedInput: 0,
5142
+ cacheRead: 0,
5143
+ cacheWrite: 0,
5144
+ output: 0
5145
+ };
5146
+ const rawCostByPricing = /* @__PURE__ */ new Map();
5147
+ const native = nativeClaudeListWriteContext(settings, model);
5148
+ for (const entry of entries) {
5149
+ if (!knownTokenCounter(entry.inputTokens) || !knownTokenCounter(entry.outputTokens))
5150
+ return unknown;
5151
+ const cached = knownDetailTokenCounter(entry, [
5152
+ "cached_tokens",
5153
+ "cachedInputTokens",
5154
+ "cached_input_tokens"
5155
+ ]);
5156
+ const writes = knownDetailTokenCounter(entry, ["cache_write_tokens", "cacheWriteTokens"], true);
5157
+ if (cached === void 0 || writes === void 0 || cached + writes > entry.inputTokens)
5158
+ return unknown;
5159
+ const pricing = selectModelPricing(schedule, entry.inputTokens);
5160
+ if (cached > 0 && pricing.cachedInputMicrosPerMillionTokens === void 0) return unknown;
5161
+ if (writes > 0 && pricing.cacheWriteMicrosPerMillionTokens === void 0) return unknown;
5162
+ const costs = {
5163
+ uncachedInput: Math.ceil(
5164
+ (entry.inputTokens - cached - writes) * pricing.inputMicrosPerMillionTokens / 1e6
5165
+ ),
5166
+ cacheRead: Math.ceil(cached * (pricing.cachedInputMicrosPerMillionTokens ?? 0) / 1e6),
5167
+ cacheWrite: Math.ceil(writes * (pricing.cacheWriteMicrosPerMillionTokens ?? 0) / 1e6),
5168
+ output: Math.ceil(entry.outputTokens * pricing.outputMicrosPerMillionTokens / 1e6)
5169
+ };
5170
+ if (native && writes > 0) {
5171
+ const fiveMinute = knownDetailTokenCounter(entry, ["cache_write_tokens_5m"]);
5172
+ const oneHour = knownDetailTokenCounter(entry, ["cache_write_tokens_1h"]);
5173
+ if (fiveMinute === void 0 || oneHour === void 0 || fiveMinute + oneHour !== writes)
5174
+ return unknown;
5175
+ if (native.publishedRates) {
5176
+ costs.cacheWrite = Math.ceil(fiveMinute * native.publishedRates.fiveMinute / 1e6) + Math.ceil(oneHour * native.publishedRates.oneHour / 1e6);
5177
+ } else if (!(native.declaredTtl === "5m" && oneHour === 0 || native.declaredTtl === "1h" && fiveMinute === 0)) {
5178
+ return unknown;
5179
+ }
5180
+ }
5181
+ const raw = sumListCostClasses(costs);
5182
+ rawCostByPricing.set(pricing, (rawCostByPricing.get(pricing) ?? 0) + raw);
5183
+ for (const key of MODEL_LIST_COST_CLASS_KEYS) classes[key] += costs[key];
5184
+ }
5185
+ let providerCostMicros = sumListCostClasses(classes);
5186
+ let creditCostMicros = 0;
5187
+ for (const [pricing, raw] of rawCostByPricing) {
5188
+ creditCostMicros += Math.ceil(raw * (1e4 + (pricing.marginBps ?? 0)) / 1e4);
5189
+ }
5190
+ let listByClassApprox = false;
5191
+ const latencyMode = options.latencyMode ?? "standard";
5192
+ if (latencyMode !== "standard") {
5193
+ const catalogSettings = settingsForTurnExecutionPolicy(settings, model);
5194
+ const resolved = resolveModelProvider(
5195
+ catalogSettings,
5196
+ canonicalizeConfiguredModelId(catalogSettings, model)
5197
+ );
5198
+ const multiplier = resolved?.model.capabilities.latencyModes.find(
5199
+ (mode) => mode.id === latencyMode && mode.runnable
5200
+ )?.billingMultiplierBps;
5201
+ if (multiplier === void 0 || multiplier <= 0) return unknown;
5202
+ providerCostMicros = Math.ceil(providerCostMicros * multiplier / 1e4);
5203
+ creditCostMicros = Math.ceil(creditCostMicros * multiplier / 1e4);
5204
+ const scaled = allocateLatencyListCostClasses(classes, multiplier, providerCostMicros);
5205
+ if (!scaled) return unknown;
5206
+ Object.assign(classes, scaled.classes);
5207
+ listByClassApprox = scaled.approximate;
5208
+ }
5209
+ if (![providerCostMicros, creditCostMicros, ...Object.values(classes)].every(knownTokenCounter))
5210
+ return unknown;
5211
+ return { providerCostMicros, creditCostMicros, listByClassMicros: classes, listByClassApprox };
5212
+ }
5213
+ function allocateRecordedModelListCostByClass(settings, model, usage, recordedProviderCostMicros) {
5214
+ const unknown = {
5215
+ listByClassMicros: null,
5216
+ listByClassApprox: false
5217
+ };
5218
+ if (!knownTokenCounter(recordedProviderCostMicros)) return unknown;
5219
+ const schedule = configuredModelListPricingSchedules(settings)[model];
5220
+ if (!schedule) return unknown;
5221
+ const weights = { uncachedInput: 0n, cacheRead: 0n, cacheWrite: 0n, output: 0n };
5222
+ const entries = usage.requestUsageEntries?.length ? usage.requestUsageEntries : [usage];
5223
+ for (const entry of entries) {
5224
+ if (!knownTokenCounter(entry.inputTokens) || !knownTokenCounter(entry.outputTokens))
5225
+ return unknown;
5226
+ const cached = knownDetailTokenCounter(entry, [
5227
+ "cached_tokens",
5228
+ "cachedInputTokens",
5229
+ "cached_input_tokens"
5230
+ ]);
5231
+ const writes = knownDetailTokenCounter(entry, ["cache_write_tokens", "cacheWriteTokens"], true);
5232
+ if (cached === void 0 || writes === void 0 || cached + writes > entry.inputTokens)
5233
+ return unknown;
5234
+ const pricing = selectModelPricing(schedule, entry.inputTokens);
5235
+ const counters = {
5236
+ uncachedInput: entry.inputTokens - cached - writes,
5237
+ cacheRead: cached,
5238
+ cacheWrite: writes,
5239
+ output: entry.outputTokens
5240
+ };
5241
+ const rates = {
5242
+ uncachedInput: pricing.inputMicrosPerMillionTokens,
5243
+ cacheRead: pricing.cachedInputMicrosPerMillionTokens,
5244
+ cacheWrite: pricing.cacheWriteMicrosPerMillionTokens,
5245
+ output: pricing.outputMicrosPerMillionTokens
5246
+ };
5247
+ for (const key of MODEL_LIST_COST_CLASS_KEYS) {
5248
+ if (counters[key] === 0) continue;
5249
+ const rate = rates[key];
5250
+ if (!knownTokenCounter(rate)) return unknown;
5251
+ weights[key] += BigInt(counters[key]) * BigInt(rate);
5252
+ }
5253
+ }
5254
+ const classes = allocateRecordedListCostWeights(weights, recordedProviderCostMicros);
5255
+ return classes ? { listByClassMicros: classes, listByClassApprox: true } : unknown;
5256
+ }
5257
+ var MODEL_LIST_COST_CLASS_KEYS = ["uncachedInput", "cacheRead", "cacheWrite", "output"];
5258
+ function sumListCostClasses(costs) {
5259
+ return MODEL_LIST_COST_CLASS_KEYS.reduce((total, key) => total + costs[key], 0);
5260
+ }
5261
+ function allocateRecordedListCostWeights(weights, target) {
5262
+ const denominator = MODEL_LIST_COST_CLASS_KEYS.reduce((total, key) => total + weights[key], 0n);
5263
+ if (denominator === 0n) {
5264
+ return target === 0 ? { uncachedInput: 0, cacheRead: 0, cacheWrite: 0, output: 0 } : void 0;
5265
+ }
5266
+ const rows = MODEL_LIST_COST_CLASS_KEYS.map((key, index) => {
5267
+ const numerator = BigInt(target) * weights[key];
5268
+ return { key, index, value: numerator / denominator, remainder: numerator % denominator };
5269
+ });
5270
+ const remaining = BigInt(target) - rows.reduce((total, row) => total + row.value, 0n);
5271
+ const ranked = [...rows].sort(
5272
+ (a, b) => a.remainder === b.remainder ? a.index - b.index : a.remainder > b.remainder ? -1 : 1
5273
+ );
5274
+ for (let index = 0; index < Number(remaining); index++) ranked[index].value += 1n;
5275
+ return Object.fromEntries(
5276
+ rows.map((row) => [row.key, Number(row.value)])
5277
+ );
5278
+ }
5279
+ function knownTokenCounter(value) {
5280
+ return typeof value === "number" && Number.isSafeInteger(value) && value >= 0;
5281
+ }
5282
+ function knownDetailTokenCounter(entry, keys, firstOnly = false) {
5283
+ const details = Array.isArray(entry.inputTokensDetails) ? entry.inputTokensDetails : entry.inputTokensDetails ? [entry.inputTokensDetails] : [];
5284
+ if (details.length === 0) return void 0;
5285
+ let total = 0;
5286
+ for (const detail of details) {
5287
+ const present = keys.filter((key) => detail[key] !== void 0);
5288
+ if (present.length === 0) return void 0;
5289
+ for (const key of firstOnly ? present.slice(0, 1) : present) {
5290
+ if (!knownTokenCounter(detail[key])) return void 0;
5291
+ total += detail[key];
5292
+ }
5293
+ }
5294
+ return knownTokenCounter(total) ? total : void 0;
5295
+ }
5296
+ function nativeClaudeListWriteContext(settings, modelId) {
5297
+ for (const provider of configuredRegistryProviders(settings)) {
5298
+ const model = provider.models.find((candidate) => candidate.id === modelId);
5299
+ if (!model || provider.api !== "anthropic-messages") continue;
5300
+ const result = { declaredTtl: provider.anthropic?.cacheTtl };
5301
+ const upstream = model.upstreamModelId ?? model.id;
5302
+ const overrides = parseModelPricingJson(settings.modelPricingJson);
5303
+ const official = provider.kind === "anthropic-workspace" || provider.kind === "anthropic-organization" || provider.kind === "claude-subscription-workspace" || provider.kind === "claude-subscription-organization" || provider.kind === "api-key" && provider.baseUrl.replace(/\/$/u, "") === "https://api.anthropic.com/v1";
5304
+ const published = reviewedModelListPricing[upstream]?.default;
5305
+ if (!official || !claudeNativeModelProfile(upstream) || !published?.cacheWriteMicrosPerMillionTokens || model.pricing || overrides[modelId] !== void 0 || overrides[upstream] !== void 0)
5306
+ return result;
5307
+ return {
5308
+ ...result,
5309
+ publishedRates: {
5310
+ fiveMinute: published.cacheWriteMicrosPerMillionTokens,
5311
+ oneHour: published.inputMicrosPerMillionTokens * 2
5312
+ }
5313
+ };
5314
+ }
5315
+ return void 0;
5316
+ }
5317
+ function allocateLatencyListCostClasses(classes, multiplier, target) {
5318
+ if (!knownTokenCounter(multiplier) || !knownTokenCounter(target) || !Object.values(classes).every(knownTokenCounter))
5319
+ return void 0;
5320
+ const rows = MODEL_LIST_COST_CLASS_KEYS.map((key, index) => {
5321
+ const numerator = BigInt(classes[key]) * BigInt(multiplier);
5322
+ return { key, index, value: Number(numerator / 10000n), remainder: numerator % 10000n };
5323
+ });
5324
+ const remaining = target - rows.reduce((total, row) => total + row.value, 0);
5325
+ if (remaining < 0 || remaining > rows.length) return void 0;
5326
+ const ranked = [...rows].sort(
5327
+ (a, b) => a.remainder === b.remainder ? a.index - b.index : a.remainder > b.remainder ? -1 : 1
5328
+ );
5329
+ for (let index = 0; index < remaining; index++) ranked[index].value += 1;
5330
+ return {
5331
+ classes: Object.fromEntries(
5332
+ rows.map((row) => [row.key, row.value])
5333
+ ),
5334
+ approximate: rows.some((row) => row.remainder !== 0n)
5335
+ };
5336
+ }
5337
+ function calculateUsageCostBreakdown(settings, model, usage, schedule, options) {
3964
5338
  if (!schedule) {
3965
5339
  throw new Error(`Missing model pricing for ${model}`);
3966
5340
  }
@@ -4219,7 +5593,7 @@ async function retryStartupDependency(label, operation, options = {}) {
4219
5593
  try {
4220
5594
  return await operation();
4221
5595
  } catch (error) {
4222
- if (attempt >= attempts) {
5596
+ if (attempt >= attempts || options.shouldRetry?.(error) === false) {
4223
5597
  throw error;
4224
5598
  }
4225
5599
  const delayMs = Math.min(maxDelayMs, initialDelayMs * 2 ** (attempt - 1));
@@ -4309,7 +5683,7 @@ function parseModelPricingJson(raw) {
4309
5683
  if (!model.trim()) {
4310
5684
  throw new Error("OPENGENI_MODEL_PRICING_JSON contains an empty model name");
4311
5685
  }
4312
- out[model] = z.union([ModelPricingSchema, ModelPricingScheduleSchema]).parse(value);
5686
+ out[model] = z2.union([ModelPricingSchema, ModelPricingScheduleSchema]).parse(value);
4313
5687
  }
4314
5688
  return out;
4315
5689
  }
@@ -4419,9 +5793,9 @@ function parseIntegrationsOauthClientsJson(raw) {
4419
5793
  }
4420
5794
  return out;
4421
5795
  }
4422
- var SocialOAuthClientConfigSchema = z.object({
4423
- clientId: z.string().min(1),
4424
- clientSecret: z.string().min(1).optional()
5796
+ var SocialOAuthClientConfigSchema = z2.object({
5797
+ clientId: z2.string().min(1),
5798
+ clientSecret: z2.string().min(1).optional()
4425
5799
  });
4426
5800
  var SOCIAL_OAUTH_PROVIDER_IDS = ["x", "reddit"];
4427
5801
  function parseSocialOauthClientsJson(raw) {
@@ -4558,20 +5932,9 @@ function ensureBuiltInMcpServers(settings) {
4558
5932
  ...hasDocs ? [] : [
4559
5933
  {
4560
5934
  id: "docs",
4561
- name: "Document Search",
5935
+ name: "Knowledge",
4562
5936
  url: firstPartyDocsMcpUrl,
4563
- allowedTools: [
4564
- "search_documents",
4565
- "fetch_document_chunk",
4566
- "list_document_bases",
4567
- "list_indexed_documents",
4568
- "knowledge_search",
4569
- "knowledge_get",
4570
- "knowledge_browse",
4571
- "knowledge_fetch",
4572
- "memory_search",
4573
- "memory_propose"
4574
- ],
5937
+ allowedTools: ["knowledge_search", "knowledge_get", "knowledge_browse"],
4575
5938
  cacheToolsList: false
4576
5939
  }
4577
5940
  ],
@@ -4630,8 +5993,61 @@ function isDigestPinnedModalDesktopImage(settings) {
4630
5993
  if (settings.modalImageId) return true;
4631
5994
  return typeof settings.modalImageRef === "string" && MODAL_DESKTOP_IMAGE_DIGEST_REF.test(settings.modalImageRef);
4632
5995
  }
5996
+ function trustedProxyCidrEntries(raw) {
5997
+ const entries = [];
5998
+ for (const entry of raw.split(",")) {
5999
+ const value = entry.trim();
6000
+ if (!value) continue;
6001
+ const slash = value.indexOf("/");
6002
+ const address = slash < 0 ? value : value.slice(0, slash);
6003
+ const version = isIP(address);
6004
+ const family = version === 4 ? "ipv4" : version === 6 ? "ipv6" : null;
6005
+ const maxPrefix = family === "ipv4" ? 32 : 128;
6006
+ const prefixText = slash < 0 ? String(maxPrefix) : value.slice(slash + 1);
6007
+ const prefix = /^\d{1,3}$/u.test(prefixText) ? Number(prefixText) : Number.NaN;
6008
+ if (!family || !Number.isInteger(prefix) || prefix < 0 || prefix > maxPrefix) {
6009
+ throw new Error(
6010
+ `OPENGENI_API_TRUSTED_PROXY_CIDRS entry "${value.slice(0, 64)}" is not an IP address or CIDR range`
6011
+ );
6012
+ }
6013
+ entries.push({ address, prefix, family });
6014
+ }
6015
+ return entries;
6016
+ }
6017
+ function localAllowedOriginEntries(raw) {
6018
+ const origins = [];
6019
+ for (const entry of (raw ?? "").split(",")) {
6020
+ const value = entry.trim();
6021
+ if (!value) continue;
6022
+ let url = null;
6023
+ try {
6024
+ url = new URL(value);
6025
+ } catch {
6026
+ url = null;
6027
+ }
6028
+ if (!url || url.protocol !== "http:" && url.protocol !== "https:" || url.username || url.password || url.hostname.includes("*") || url.pathname !== "/" || url.search || url.hash || value.endsWith("?") || value.endsWith("#")) {
6029
+ throw new Error(
6030
+ `OPENGENI_LOCAL_ALLOWED_ORIGINS entry "${value.slice(0, 64)}" is not an exact http(s) origin such as http://127.0.0.1:5173`
6031
+ );
6032
+ }
6033
+ origins.push(url.origin);
6034
+ }
6035
+ return origins;
6036
+ }
6037
+ var LOCAL_SANDBOX_API_ROUTES = [
6038
+ // Codemode calls, journal reads, catalog, and the Site/SDK proxy.
6039
+ /^\/v1\/workspaces\/[^/]+\/codemode(?:\/|$)/u,
6040
+ // First-party MCP, including its /docs and /files servers.
6041
+ /^\/v1\/workspaces\/[^/]+\/mcp(?:\/|$)/u,
6042
+ // The personal GitHub HTTPS smart-Git broker.
6043
+ /^\/v1\/git\/personal\/[A-Za-z0-9_-]{43}\/(?:info\/refs|git-upload-pack|git-receive-pack)$/u
6044
+ ];
6045
+ function localSandboxApiRouteAllowed(pathname) {
6046
+ return LOCAL_SANDBOX_API_ROUTES.some((route) => route.test(pathname));
6047
+ }
4633
6048
  function validateSettings(settings, source = process.env) {
4634
6049
  temporalConnectionOptions(settings);
6050
+ localAllowedOriginEntries(settings.localAllowedOrigins);
4635
6051
  if (settings.organizationUserSetupEmailTokenTransport === "query" && !settings.organizationUserSetupQueryEdgeSanitizationConfirmed) {
4636
6052
  throw new Error(
4637
6053
  "OPENGENI_ORGANIZATION_USER_SETUP_QUERY_EDGE_SANITIZATION_CONFIRMED=true is required when OPENGENI_ORGANIZATION_USER_SETUP_EMAIL_TOKEN_TRANSPORT=query"
@@ -4714,6 +6130,18 @@ function validateSettings(settings, source = process.env) {
4714
6130
  );
4715
6131
  }
4716
6132
  }
6133
+ const retiredTrustedProxyHops = source.OPENGENI_MCP_OAUTH_TRUSTED_PROXY_HOPS?.trim();
6134
+ if (retiredTrustedProxyHops && retiredTrustedProxyHops !== "0") {
6135
+ throw new Error(
6136
+ "OPENGENI_MCP_OAUTH_TRUSTED_PROXY_HOPS was renamed to OPENGENI_API_TRUSTED_PROXY_HOPS, which now sets the client address for every API rate limit and auth session; rename the variable"
6137
+ );
6138
+ }
6139
+ const trustedProxyCidrs = trustedProxyCidrEntries(settings.apiTrustedProxyCidrs);
6140
+ if (trustedProxyCidrs.length > 0 && settings.apiTrustedProxyHops === 0) {
6141
+ throw new Error(
6142
+ "OPENGENI_API_TRUSTED_PROXY_CIDRS requires OPENGENI_API_TRUSTED_PROXY_HOPS greater than 0"
6143
+ );
6144
+ }
4717
6145
  if (settings.mcpOauthEnabled) {
4718
6146
  if (settings.productAccessMode === "configured") {
4719
6147
  throw new Error(
@@ -4974,60 +6402,7 @@ function validateSettings(settings, source = process.env) {
4974
6402
  );
4975
6403
  }
4976
6404
  }
4977
- if (settings.objectStorageBackend === "s3-compatible" || settings.objectStorageBackend === "aws-s3") {
4978
- if (Boolean(settings.objectStorageAccessKeyId) !== Boolean(settings.objectStorageSecretAccessKey)) {
4979
- throw new Error(
4980
- "OPENGENI_OBJECT_STORAGE_ACCESS_KEY_ID and OPENGENI_OBJECT_STORAGE_SECRET_ACCESS_KEY must both be set or both omitted"
4981
- );
4982
- }
4983
- if (settings.objectStorageBackend === "s3-compatible" && (settings.objectStorageEndpoint || settings.objectStorageInternalEndpoint || settings.objectStorageSandboxEndpoint) && (!settings.objectStorageAccessKeyId || !settings.objectStorageSecretAccessKey)) {
4984
- throw new Error(
4985
- "S3-compatible object storage endpoints require OPENGENI_OBJECT_STORAGE_ACCESS_KEY_ID and OPENGENI_OBJECT_STORAGE_SECRET_ACCESS_KEY"
4986
- );
4987
- }
4988
- if (settings.objectStorageAzureConnectionString || settings.objectStorageAzureAccountName || settings.objectStorageAzureAccountKey || settings.objectStorageAzureEndpoint) {
4989
- throw new Error(
4990
- "S3 object storage uses OPENGENI_OBJECT_STORAGE_* S3 settings, not OPENGENI_OBJECT_STORAGE_AZURE_* settings"
4991
- );
4992
- }
4993
- if (settings.objectStorageGcsProjectId || settings.objectStorageGcsCredentialsJson || settings.objectStorageGcsKeyFilename || settings.objectStorageGcsApiEndpoint) {
4994
- throw new Error(
4995
- "S3 object storage uses OPENGENI_OBJECT_STORAGE_* S3 settings, not OPENGENI_OBJECT_STORAGE_GCS_* settings"
4996
- );
4997
- }
4998
- } else if (settings.objectStorageBackend === "azure-blob") {
4999
- if (settings.objectStorageEndpoint || settings.objectStorageInternalEndpoint || settings.objectStorageSandboxEndpoint || settings.objectStorageAccessKeyId || settings.objectStorageSecretAccessKey) {
5000
- throw new Error(
5001
- "Azure Blob storage uses OPENGENI_OBJECT_STORAGE_AZURE_* settings, not S3-compatible object storage settings"
5002
- );
5003
- }
5004
- if (settings.objectStorageGcsProjectId || settings.objectStorageGcsCredentialsJson || settings.objectStorageGcsKeyFilename || settings.objectStorageGcsApiEndpoint) {
5005
- throw new Error(
5006
- "Azure Blob storage uses OPENGENI_OBJECT_STORAGE_AZURE_* settings, not OPENGENI_OBJECT_STORAGE_GCS_* settings"
5007
- );
5008
- }
5009
- const hasConnectionString = Boolean(settings.objectStorageAzureConnectionString);
5010
- const hasSharedKey = Boolean(settings.objectStorageAzureAccountName) && Boolean(settings.objectStorageAzureAccountKey);
5011
- if (!hasConnectionString && !hasSharedKey) {
5012
- throw new Error(
5013
- "Azure Blob storage requires OPENGENI_OBJECT_STORAGE_AZURE_CONNECTION_STRING or OPENGENI_OBJECT_STORAGE_AZURE_ACCOUNT_NAME plus OPENGENI_OBJECT_STORAGE_AZURE_ACCOUNT_KEY"
5014
- );
5015
- }
5016
- } else {
5017
- if (settings.objectStorageEndpoint || settings.objectStorageInternalEndpoint || settings.objectStorageSandboxEndpoint || settings.objectStorageAccessKeyId || settings.objectStorageSecretAccessKey) {
5018
- throw new Error(
5019
- "GCS object storage uses OPENGENI_OBJECT_STORAGE_GCS_* settings, not S3-compatible object storage settings"
5020
- );
5021
- }
5022
- if (settings.objectStorageAzureConnectionString || settings.objectStorageAzureAccountName || settings.objectStorageAzureAccountKey || settings.objectStorageAzureEndpoint) {
5023
- throw new Error(
5024
- "GCS object storage uses OPENGENI_OBJECT_STORAGE_GCS_* settings, not OPENGENI_OBJECT_STORAGE_AZURE_* settings"
5025
- );
5026
- }
5027
- if (settings.objectStorageGcsCredentialsJson) {
5028
- parseGcsCredentialsJson(settings.objectStorageGcsCredentialsJson);
5029
- }
5030
- }
6405
+ validateObjectStorageSettings(settings);
5031
6406
  if (settings.documentChunkOverlap >= settings.documentChunkSize) {
5032
6407
  throw new Error(
5033
6408
  "OPENGENI_DOCUMENT_CHUNK_OVERLAP must be smaller than OPENGENI_DOCUMENT_CHUNK_SIZE"
@@ -5036,7 +6411,22 @@ function validateSettings(settings, source = process.env) {
5036
6411
  parseExposedPorts(settings.dockerExposedPorts);
5037
6412
  sandboxEnvironmentVariableNames(settings);
5038
6413
  sandboxLifecycleHookIds(settings);
5039
- parseSandboxWarmRateJson(settings.sandboxWarmRateMicrosPerSecondJson);
6414
+ const warmRates = parseSandboxWarmRateJson(settings.sandboxWarmRateMicrosPerSecondJson);
6415
+ if (settings.sandboxWarmBillingMode === "credits" && Object.values(warmRates).some((rate) => !Number.isSafeInteger(rate))) {
6416
+ throw new Error(
6417
+ "OPENGENI_SANDBOX_WARM_RATE_MICROS_PER_SECOND_JSON paid rates must be safe integers"
6418
+ );
6419
+ }
6420
+ if (settings.documentEmbeddingBillingMode === "credits" && settings.documentEmbeddingProvider === "openai" && settings.documentEmbeddingRateMicrosPerMillionBytes <= 0) {
6421
+ throw new Error(
6422
+ "OPENGENI_DOCUMENT_EMBEDDING_RATE_MICROS_PER_MILLION_BYTES must be positive when paid OpenAI embedding billing is enabled"
6423
+ );
6424
+ }
6425
+ if (settings.documentEmbeddingBillingMode === "credits" && settings.documentEmbeddingProvider === "openai" && !settings.documentEmbeddingCreditsActivatedAt) {
6426
+ throw new Error(
6427
+ "OPENGENI_DOCUMENT_EMBEDDING_CREDITS_ACTIVATED_AT is required for paid OpenAI embeddings"
6428
+ );
6429
+ }
5040
6430
  if (settings.sandboxBackend === "opensandbox") {
5041
6431
  if (!/@sha256:[0-9a-f]{64}$/i.test(settings.openSandboxImage ?? "")) {
5042
6432
  throw new Error(
@@ -5055,9 +6445,9 @@ function validateSettings(settings, source = process.env) {
5055
6445
  throw new Error(`OPENGENI_MCP_SERVERS contains duplicate id ${server.id}`);
5056
6446
  }
5057
6447
  serverIds.add(server.id);
5058
- if (server.connectionRef?.authoritySource === "host" && !settings.hostMcpAuthoritySourceAdmissionEnabled) {
6448
+ if (server.connectionRef?.authoritySource === "host") {
5059
6449
  throw new Error(
5060
- "OPENGENI_MCP_SERVERS host-owned connection refs require OPENGENI_HOST_MCP_AUTHORITY_SOURCE_ADMISSION_ENABLED=true after the whole API/worker fleet is upgraded"
6450
+ "OPENGENI_MCP_SERVERS host-owned connection refs are no longer supported; select ordinary native connections"
5061
6451
  );
5062
6452
  }
5063
6453
  }
@@ -5095,9 +6485,21 @@ function validateSettings(settings, source = process.env) {
5095
6485
  ordinaryCaptureTimeoutMs,
5096
6486
  drainCaptureTimeoutMs
5097
6487
  );
5098
- if (!(rotationLeadMs > providerDeadlineCaptureTimeoutMs + reaperPeriod)) {
6488
+ const requiredRotationLeadMs = SANDBOX_DEADLINE_COMMAND_STOP_GRACE_MS + providerDeadlineCaptureTimeoutMs + 2 * reaperPeriod;
6489
+ if (!(rotationLeadMs > requiredRotationLeadMs)) {
6490
+ throw new Error(
6491
+ `OPENGENI_SANDBOX_ROTATION_LEAD_MS (${rotationLeadMs}) must exceed the legacy command stop grace, largest durable snapshot or drain capture timeout, and two reaper periods (${requiredRotationLeadMs}), including for persisted Modal leases after a default-backend rollout.`
6492
+ );
6493
+ }
6494
+ const containmentMs = settings.sandboxIdleCommandContainmentMs;
6495
+ if (containmentMs !== void 0 && !(containmentMs > settings.sandboxIdleGraceMs)) {
6496
+ throw new Error(
6497
+ `OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS (${containmentMs}) must exceed OPENGENI_SANDBOX_IDLE_GRACE_MS (${settings.sandboxIdleGraceMs}): a box kept warm only by retained commands must stay available at least as long as an idle box awaiting drain.`
6498
+ );
6499
+ }
6500
+ if (containmentMs !== void 0 && optionalEnvironmentValue("OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS", source) !== void 0 && !(containmentMs < rotationLeadMs)) {
5099
6501
  throw new Error(
5100
- `OPENGENI_SANDBOX_ROTATION_LEAD_MS (${rotationLeadMs}) must exceed the largest durable snapshot or drain capture timeout plus one reaper period (${providerDeadlineCaptureTimeoutMs + reaperPeriod}), including for persisted Modal leases after a default-backend rollout.`
6502
+ `OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS (${containmentMs}) must be strictly less than OPENGENI_SANDBOX_ROTATION_LEAD_MS (${rotationLeadMs}): an idle box must be checkpointed and stopped well before its provider deadline.`
5101
6503
  );
5102
6504
  }
5103
6505
  if (settings.sandboxBackend === "modal") {
@@ -5130,6 +6532,11 @@ function validateSettings(settings, source = process.env) {
5130
6532
  `OPENGENI_SANDBOX_LEASE_REAPER_PERIOD_MS + OPENGENI_SANDBOX_IDLE_GRACE_MS (${reaperPeriod} + ${idleGraceMs} = ${reaperPeriod + idleGraceMs}) must be strictly less than the effective box idle timeout (${idleTimeoutMs}): a drained box must SURVIVE its full warm window so the reaper can resume + snapshot /workspace + terminate it on the sweep AFTER the drain grace elapses \u2014 Modal's idle-reap must NOT fire first (or /workspace is lost). Raise OPENGENI_MODAL_IDLE_TIMEOUT_SECONDS (defaults to OPENGENI_MODAL_TIMEOUT_SECONDS) or lower OPENGENI_SANDBOX_IDLE_GRACE_MS.`
5131
6533
  );
5132
6534
  }
6535
+ if (settings.modalIdleTimeoutSeconds !== void 0 && containmentMs !== void 0 && !(reaperPeriod + containmentMs + drainCaptureTimeoutMs < idleTimeoutMs)) {
6536
+ throw new Error(
6537
+ `OPENGENI_SANDBOX_LEASE_REAPER_PERIOD_MS + OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS + drain capture timeout (${reaperPeriod} + ${containmentMs} + ${drainCaptureTimeoutMs}) must be strictly less than OPENGENI_MODAL_IDLE_TIMEOUT_SECONDS*1000 (${idleTimeoutMs}): idle command containment must checkpoint a box before Modal's own idle reaper stops it. Lower OPENGENI_SANDBOX_IDLE_COMMAND_CONTAINMENT_MS or raise the Modal idle timeout.`
6538
+ );
6539
+ }
5133
6540
  }
5134
6541
  }
5135
6542
  if (settings.sandboxDesktopEnabled && settings.sandboxBackend === "modal" && !["local", "test"].includes(settings.environment)) {
@@ -5157,7 +6564,7 @@ function validateModelCatalogSettings(settings, source = process.env) {
5157
6564
  const builtinId = builtinProviderId(settings);
5158
6565
  const providerIds = /* @__PURE__ */ new Set();
5159
6566
  for (const provider of registryProviders) {
5160
- if (provider.kind === "vercel-gateway-managed" || provider.kind === "vercel-gateway-workspace" || provider.kind === "vercel-gateway-organization" || provider.kind === "openrouter-workspace" || provider.kind === "openrouter-organization" || provider.kind === "xai-subscription") {
6567
+ if (provider.kind === "vercel-gateway-managed" || provider.kind === "vercel-gateway-workspace" || provider.kind === "vercel-gateway-organization" || provider.kind === "direct-openai-workspace" || provider.kind === "direct-azure-workspace" || provider.kind === "openrouter-workspace" || provider.kind === "openrouter-organization" || provider.kind === "anthropic-organization" || provider.kind === "anthropic-workspace" || provider.kind === "claude-subscription-workspace" || provider.kind === "claude-subscription-organization" || provider.kind === "xai-subscription") {
5161
6568
  throw new Error(
5162
6569
  `OPENGENI_MODEL_PROVIDERS_JSON provider kind ${provider.kind} is reserved for a reviewed OpenGeni credential broker`
5163
6570
  );
@@ -5199,10 +6606,25 @@ function validateModelCatalogSettings(settings, source = process.env) {
5199
6606
  `The default model ${settings.openaiModel} is not executable in the resolved model catalog`
5200
6607
  );
5201
6608
  }
6609
+ const explicitCreditsDefault = source.OPENGENI_CREDITS_DEFAULT_MODEL?.trim();
6610
+ if (explicitCreditsDefault && settings.modelCatalogSource === "code" && settings.billingMode === "stripe") {
6611
+ const creditsDefaultId = canonicalizeConfiguredModelId(settings, settings.creditsDefaultModel);
6612
+ const creditsDefault = models.find((model) => model.id === creditsDefaultId);
6613
+ if (creditsDefault?.cost !== "credits") {
6614
+ throw new Error(
6615
+ `OPENGENI_CREDITS_DEFAULT_MODEL ${settings.creditsDefaultModel} is not a credits-billed model in the resolved model catalog`
6616
+ );
6617
+ }
6618
+ }
5202
6619
  const deploymentProductIds = new Set(
5203
6620
  models.filter((model) => model.credentialSource.kind === "deployment").map((model) => model.id)
5204
6621
  );
5205
6622
  const noteProductIds = new Set(models.map((model) => model.id));
6623
+ if (settings.resolvedCodexModelsJson !== void 0) {
6624
+ for (const model of z2.array(CodexCatalogModelSchema).parse(JSON.parse(settings.resolvedCodexModelsJson))) {
6625
+ noteProductIds.add(model.id);
6626
+ }
6627
+ }
5206
6628
  for (const model of configuredGatewayCatalogModels(settings)) {
5207
6629
  deploymentProductIds.add(model.productId);
5208
6630
  noteProductIds.add(model.productId);
@@ -5280,6 +6702,62 @@ function resolveNatsCalloutConfig(settings) {
5280
6702
  }
5281
6703
  return { accountSeed, accountName, user, password };
5282
6704
  }
6705
+ function validateObjectStorageSettings(settings) {
6706
+ if (settings.objectStorageBackend === "s3-compatible" || settings.objectStorageBackend === "aws-s3") {
6707
+ if (Boolean(settings.objectStorageAccessKeyId) !== Boolean(settings.objectStorageSecretAccessKey)) {
6708
+ throw new Error(
6709
+ "OPENGENI_OBJECT_STORAGE_ACCESS_KEY_ID and OPENGENI_OBJECT_STORAGE_SECRET_ACCESS_KEY must both be set or both omitted"
6710
+ );
6711
+ }
6712
+ if (settings.objectStorageBackend === "s3-compatible" && (settings.objectStorageEndpoint || settings.objectStorageInternalEndpoint || settings.objectStorageSandboxEndpoint) && (!settings.objectStorageAccessKeyId || !settings.objectStorageSecretAccessKey)) {
6713
+ throw new Error(
6714
+ "S3-compatible object storage endpoints require OPENGENI_OBJECT_STORAGE_ACCESS_KEY_ID and OPENGENI_OBJECT_STORAGE_SECRET_ACCESS_KEY"
6715
+ );
6716
+ }
6717
+ if (settings.objectStorageAzureConnectionString || settings.objectStorageAzureAccountName || settings.objectStorageAzureAccountKey || settings.objectStorageAzureEndpoint) {
6718
+ throw new Error(
6719
+ "S3 object storage uses OPENGENI_OBJECT_STORAGE_* S3 settings, not OPENGENI_OBJECT_STORAGE_AZURE_* settings"
6720
+ );
6721
+ }
6722
+ if (settings.objectStorageGcsProjectId || settings.objectStorageGcsCredentialsJson || settings.objectStorageGcsKeyFilename || settings.objectStorageGcsApiEndpoint) {
6723
+ throw new Error(
6724
+ "S3 object storage uses OPENGENI_OBJECT_STORAGE_* S3 settings, not OPENGENI_OBJECT_STORAGE_GCS_* settings"
6725
+ );
6726
+ }
6727
+ } else if (settings.objectStorageBackend === "azure-blob") {
6728
+ if (settings.objectStorageEndpoint || settings.objectStorageInternalEndpoint || settings.objectStorageSandboxEndpoint || settings.objectStorageAccessKeyId || settings.objectStorageSecretAccessKey) {
6729
+ throw new Error(
6730
+ "Azure Blob storage uses OPENGENI_OBJECT_STORAGE_AZURE_* settings, not S3-compatible object storage settings"
6731
+ );
6732
+ }
6733
+ if (settings.objectStorageGcsProjectId || settings.objectStorageGcsCredentialsJson || settings.objectStorageGcsKeyFilename || settings.objectStorageGcsApiEndpoint) {
6734
+ throw new Error(
6735
+ "Azure Blob storage uses OPENGENI_OBJECT_STORAGE_AZURE_* settings, not OPENGENI_OBJECT_STORAGE_GCS_* settings"
6736
+ );
6737
+ }
6738
+ const hasConnectionString = Boolean(settings.objectStorageAzureConnectionString);
6739
+ const hasSharedKey = Boolean(settings.objectStorageAzureAccountName) && Boolean(settings.objectStorageAzureAccountKey);
6740
+ if (!hasConnectionString && !hasSharedKey) {
6741
+ throw new Error(
6742
+ "Azure Blob storage requires OPENGENI_OBJECT_STORAGE_AZURE_CONNECTION_STRING or OPENGENI_OBJECT_STORAGE_AZURE_ACCOUNT_NAME plus OPENGENI_OBJECT_STORAGE_AZURE_ACCOUNT_KEY"
6743
+ );
6744
+ }
6745
+ } else {
6746
+ if (settings.objectStorageEndpoint || settings.objectStorageInternalEndpoint || settings.objectStorageSandboxEndpoint || settings.objectStorageAccessKeyId || settings.objectStorageSecretAccessKey) {
6747
+ throw new Error(
6748
+ "GCS object storage uses OPENGENI_OBJECT_STORAGE_GCS_* settings, not S3-compatible object storage settings"
6749
+ );
6750
+ }
6751
+ if (settings.objectStorageAzureConnectionString || settings.objectStorageAzureAccountName || settings.objectStorageAzureAccountKey || settings.objectStorageAzureEndpoint) {
6752
+ throw new Error(
6753
+ "GCS object storage uses OPENGENI_OBJECT_STORAGE_GCS_* settings, not OPENGENI_OBJECT_STORAGE_AZURE_* settings"
6754
+ );
6755
+ }
6756
+ if (settings.objectStorageGcsCredentialsJson) {
6757
+ parseGcsCredentialsJson(settings.objectStorageGcsCredentialsJson);
6758
+ }
6759
+ }
6760
+ }
5283
6761
  function resolveNatsControlPlaneAuth(settings) {
5284
6762
  const user = settings.selfhostedNatsControlUser?.trim();
5285
6763
  const password = settings.selfhostedNatsControlPassword?.trim();
@@ -5291,6 +6769,9 @@ function resolveNatsControlPlaneAuth(settings) {
5291
6769
  function splitCsv(raw) {
5292
6770
  return raw.split(",").map((value) => value.trim()).filter(Boolean);
5293
6771
  }
6772
+ function sandboxImageAllowlist(settings) {
6773
+ return [...new Set(splitCsv(settings.sandboxImageAllowlist ?? ""))];
6774
+ }
5294
6775
  function uniqueEnvNames(raw, fieldName) {
5295
6776
  const seen = /* @__PURE__ */ new Set();
5296
6777
  const out = [];
@@ -5321,12 +6802,153 @@ function parseGcsCredentialsJson(raw) {
5321
6802
  function delay(ms) {
5322
6803
  return new Promise((resolve) => setTimeout(resolve, ms));
5323
6804
  }
6805
+ var CLAUDE_CONNECTION_KINDS = ["anthropic", "claude_subscription"];
6806
+ var CLAUDE_NATIVE_MODEL_PROFILES = Object.fromEntries([
6807
+ ...[
6808
+ ["claude-opus-5-5", "medium"],
6809
+ ["claude-sonnet-5-5", "medium"],
6810
+ ["claude-opus-5", "high"],
6811
+ ["claude-sonnet-5", "high"],
6812
+ ["claude-opus-4-8", "high"],
6813
+ ["claude-opus-4-7", "xhigh"]
6814
+ ].map(([id, defaultEffort]) => [
6815
+ id,
6816
+ {
6817
+ efforts: ["low", "medium", "high", "xhigh", "max"],
6818
+ defaultEffort,
6819
+ contextWindowTokens: 1e6,
6820
+ maxOutputTokens: 128e3
6821
+ }
6822
+ ]),
6823
+ ...["claude-opus-4-6", "claude-sonnet-4-6"].map((id) => [
6824
+ id,
6825
+ {
6826
+ efforts: ["low", "medium", "high", "max"],
6827
+ defaultEffort: "high",
6828
+ contextWindowTokens: 1e6,
6829
+ maxOutputTokens: 128e3
6830
+ }
6831
+ ]),
6832
+ [
6833
+ "claude-haiku-4-5-20251001",
6834
+ {
6835
+ efforts: [],
6836
+ defaultEffort: null,
6837
+ contextWindowTokens: 2e5,
6838
+ maxOutputTokens: 64e3
6839
+ }
6840
+ ]
6841
+ ]);
6842
+ function claudeNativeModelProfile(upstreamModelId) {
6843
+ return Object.hasOwn(CLAUDE_NATIVE_MODEL_PROFILES, upstreamModelId) ? CLAUDE_NATIVE_MODEL_PROFILES[upstreamModelId] : void 0;
6844
+ }
6845
+ function claudeProviderId(kind, scope = "organization") {
6846
+ return `${scope}-${kind === "anthropic" ? "anthropic" : "claude-subscription"}`;
6847
+ }
6848
+ function withClaudeConnectionCatalog(settings, connections, scope = "organization") {
6849
+ let providers = parseModelProvidersJson(settings.modelProvidersJson);
6850
+ for (const kind of CLAUDE_CONNECTION_KINDS) {
6851
+ if (kind === "claude_subscription" && !settings.claudeSubscriptionEnabled) {
6852
+ providers = providers.filter((provider) => provider.id !== claudeProviderId(kind, scope));
6853
+ continue;
6854
+ }
6855
+ const connection = connections[kind];
6856
+ if (!connection) continue;
6857
+ const id = claudeProviderId(kind, scope);
6858
+ providers = providers.filter((provider) => provider.id !== id);
6859
+ if (!connection.models.length) continue;
6860
+ providers.push({
6861
+ id,
6862
+ label: kind === "anthropic" ? "Anthropic API" : "Claude subscription",
6863
+ kind: kind === "anthropic" ? `anthropic-${scope}` : `claude-subscription-${scope}`,
6864
+ api: "anthropic-messages",
6865
+ wireProfile: "openai",
6866
+ baseUrl: "https://api.anthropic.com/v1",
6867
+ anthropic: {
6868
+ auth: kind === "anthropic" ? "api-key" : "oauth",
6869
+ cacheTtl: "5m",
6870
+ maxOutputTokens: 128e3,
6871
+ streamIdleTimeoutMs: 6e5
6872
+ },
6873
+ models: connection.models.map((model) => {
6874
+ const profile = claudeNativeModelProfile(model.upstreamModelId);
6875
+ const adaptiveThinking = Boolean(profile?.efforts.length);
6876
+ const contextWindowTokens = profile?.contextWindowTokens ?? 2e5;
6877
+ const outputReserve = profile?.maxOutputTokens ?? 32e3;
6878
+ return {
6879
+ contextWindowTokens,
6880
+ effectiveContextWindowTokens: contextWindowTokens - outputReserve,
6881
+ autoCompactTokenLimit: contextWindowTokens === 1e6 ? 8e5 : Math.min(15e4, contextWindowTokens - outputReserve - 18e3),
6882
+ id: id + "/" + model.upstreamModelId,
6883
+ upstreamModelId: model.upstreamModelId,
6884
+ label: model.label ?? (model.upstreamModelId === "claude-opus-5-5" ? "Claude Opus 5.5" : model.upstreamModelId === "claude-sonnet-5-5" ? "Claude Sonnet 5.5" : model.upstreamModelId),
6885
+ reasoningEffort: adaptiveThinking,
6886
+ hostedWebSearch: false,
6887
+ capabilities: {
6888
+ ...legacyModelCapabilities(settings, {
6889
+ reasoningEffort: adaptiveThinking,
6890
+ hostedWebSearch: false,
6891
+ vision: true
6892
+ }),
6893
+ inputFileMediaTypes: [],
6894
+ reasoning: {
6895
+ upstream: adaptiveThinking ? "supported" : "unknown",
6896
+ runnable: adaptiveThinking,
6897
+ efforts: [...profile?.efforts ?? []],
6898
+ defaultEffort: profile?.defaultEffort ?? null,
6899
+ required: false
6900
+ },
6901
+ functionCalling: { upstream: "supported", runnable: true },
6902
+ promptCaching: { upstream: "supported", runnable: true, mode: "automatic" }
6903
+ }
6904
+ };
6905
+ })
6906
+ });
6907
+ }
6908
+ return { ...settings, modelProvidersJson: JSON.stringify(providers) };
6909
+ }
6910
+ function withClaudeConnectionCredential(settings, kind, credential, scope = "organization", credentialBinding) {
6911
+ if (kind === "claude_subscription" && !settings.claudeSubscriptionEnabled)
6912
+ throw new Error("Claude subscriptions are not enabled on this deployment");
6913
+ if (!credential.trim()) throw new Error("Claude credential is empty");
6914
+ const bundle = kind === "claude_subscription" && credential.trim().startsWith("{") ? ClaudeSubscriptionCredential.parse(JSON.parse(credential)) : null;
6915
+ return {
6916
+ ...settings,
6917
+ modelProvidersJson: JSON.stringify(
6918
+ parseModelProvidersJson(settings.modelProvidersJson).map(
6919
+ (provider) => provider.id === claudeProviderId(kind, scope) ? {
6920
+ ...provider,
6921
+ apiKey: bundle?.token ?? credential,
6922
+ ...kind === "claude_subscription" && provider.anthropic ? {
6923
+ anthropic: {
6924
+ ...provider.anthropic,
6925
+ identity: bundle?.identity,
6926
+ credentialBinding: credentialBinding ? {
6927
+ connectionId: credentialBinding.connectionId,
6928
+ credentialVersion: credentialBinding.credentialVersion
6929
+ } : void 0
6930
+ }
6931
+ } : {}
6932
+ } : provider
6933
+ )
6934
+ )
6935
+ };
6936
+ }
5324
6937
  export {
5325
6938
  AGENT_INSTRUCTIONS_CORE_PLACEHOLDER,
5326
6939
  AI_GATEWAY_REALTIME_MODELS,
6940
+ CLAUDE_CONNECTION_KINDS,
6941
+ CLAUDE_OAUTH_AUTHORIZE_URL,
6942
+ CLAUDE_OAUTH_CLIENT_ID,
6943
+ CLAUDE_OAUTH_REDIRECT_URL,
6944
+ CLAUDE_OAUTH_SCOPES,
6945
+ CLAUDE_OAUTH_TOKEN_URL,
5327
6946
  CODEX_REALTIME_MODEL_ID,
5328
6947
  CapabilityStateV1Schema,
5329
6948
  CapabilitySupportV1,
6949
+ ClaudeOAuthTokenResponse,
6950
+ ClaudeSubscriptionCredential,
6951
+ ClaudeSubscriptionIdentity,
5330
6952
  ConfiguredModelCostClass,
5331
6953
  DEFAULT_AGENT_INSTRUCTIONS,
5332
6954
  DEFAULT_GOAL_IDLE_BACKOFF_MAX_MS,
@@ -5338,6 +6960,7 @@ export {
5338
6960
  GOOGLE_DRIVE_PROVIDER_RETRY_DELAY_MAX_MS,
5339
6961
  GatewayCatalogModel,
5340
6962
  IntegrationOAuthClientConfigSchema,
6963
+ McpOperationRecoverySchema,
5341
6964
  McpServerConnectionRefSchema,
5342
6965
  ModelCapabilitiesV1Schema,
5343
6966
  ModelCatalogDocument,
@@ -5359,12 +6982,16 @@ export {
5359
6982
  RegistryProviderKind,
5360
6983
  SANDBOX_ARCHIVE_CAPTURE_MAX_TIMEOUT_MS,
5361
6984
  SANDBOX_ARCHIVE_CAPTURE_SETTLEMENT_GRACE_MS,
6985
+ SANDBOX_DEADLINE_COMMAND_STOP_GRACE_MS,
5362
6986
  SANDBOX_LIFECYCLE_RETRY_HANDOFF_GRACE_MS,
5363
6987
  SANDBOX_LIFECYCLE_TRANSITION_MAX_WAIT_MS,
5364
6988
  SANDBOX_REQUIRED_ENV,
5365
6989
  SANDBOX_SNAPSHOT_MAX_TIMEOUT_MS,
5366
6990
  SUPERGROK_REALTIME_MODEL_ID,
5367
6991
  SocialOAuthClientConfigSchema,
6992
+ TurnExecutionPolicyDefinitionMismatchError,
6993
+ TurnExecutionPolicyModelUnavailableError,
6994
+ UnsupportedLatencyModeError,
5368
6995
  VERCEL_AI_GATEWAY_AI_SDK_BASE_URL,
5369
6996
  VERCEL_AI_GATEWAY_BASE_URL,
5370
6997
  VERCEL_AI_GATEWAY_CONNECTION_DOMAIN,
@@ -5377,6 +7004,8 @@ export {
5377
7004
  WORKSPACE_OPENROUTER_PROVIDER_ID,
5378
7005
  WORKSPACE_REALTIME_MODEL_ID_PREFIX,
5379
7006
  XAI_SUBSCRIPTION_MODEL_ID_PREFIX2 as XAI_SUBSCRIPTION_MODEL_ID_PREFIX,
7007
+ agentConfigDeploymentPolicy,
7008
+ allocateRecordedModelListCostByClass,
5380
7009
  allowedFirstPartyMcpToolsForSession,
5381
7010
  applyGitAuthPointerEnvironment,
5382
7011
  applyModelCatalogDocument,
@@ -5385,12 +7014,20 @@ export {
5385
7014
  calculateGatewayReportedCostBreakdown,
5386
7015
  calculateGatewayReportedCostMicros,
5387
7016
  calculateGatewayReportedProviderCostMicros,
7017
+ calculateModelListUsageCostBreakdown,
7018
+ calculateModelListUsageCostSnapshot,
5388
7019
  calculateModelUsageCostBreakdown,
5389
7020
  calculateModelUsageCostMicros,
5390
7021
  calculateVideoGenerationCreditCostMicros,
5391
7022
  canonicalPublicOrigin,
5392
7023
  canonicalizeConfiguredModelId,
7024
+ claudeAuthorizationUrl,
7025
+ claudeNativeModelProfile,
7026
+ claudeProviderId,
7027
+ claudeSubscriptionCapacity,
7028
+ codeSearchDeploymentPolicy,
5393
7029
  codemodeWorkspaceUrl,
7030
+ codexUpstreamModelSlugs,
5394
7031
  collectGitIdentityEnvironment,
5395
7032
  collectSandboxEnvironment,
5396
7033
  configuredAllowedModels,
@@ -5401,6 +7038,7 @@ export {
5401
7038
  configuredGatewayWorkspaceProductModelIds,
5402
7039
  configuredGoogleDriveSyncLimits,
5403
7040
  configuredModelInputIdentities,
7041
+ configuredModelListPricingSchedules,
5404
7042
  configuredModelNotes,
5405
7043
  configuredModelPricing,
5406
7044
  configuredModelPricingSchedules,
@@ -5416,20 +7054,31 @@ export {
5416
7054
  effectiveModalIdleTimeoutSeconds,
5417
7055
  effectiveSandboxDrainSnapshotTimeoutMs,
5418
7056
  effectiveSandboxLifecycle,
7057
+ emptyClaudeUsage,
5419
7058
  environmentsEncryptionKeyBytes,
5420
7059
  firstPartyMcpBaseUrl,
5421
7060
  firstPartyMcpInternalBaseUrl,
5422
7061
  firstPartyMcpInternalWorkspaceUrl,
5423
7062
  firstPartyMcpWorkspaceUrl,
5424
7063
  gatewayRequestPolicyForUpstreamModel,
7064
+ getArtifactMaterializerSettings,
7065
+ getArtifactOutboxSettings,
5425
7066
  getSettings,
5426
7067
  googleDriveOAuthCallbackUrl,
5427
7068
  googleDriveProviderRetryOptions,
5428
7069
  hasGitCredentialRepositorySelection,
5429
7070
  hasGitHubRepositorySelection,
5430
7071
  isDirectOpenAiApiBaseUrl,
7072
+ isManagedOpenRouterFreeRoute,
7073
+ isModelAvailableForNewSelection,
5431
7074
  isUsableVoiceInputSecret,
7075
+ localAllowedOriginEntries,
7076
+ localSandboxApiRouteAllowed,
7077
+ managedUserEmailAllowed,
7078
+ mergeClaudeUsage,
5432
7079
  modelCostClassForConfiguredModel,
7080
+ parseClaudeUsageHeaders,
7081
+ parseClaudeUsageResponse,
5433
7082
  parseExposedPorts,
5434
7083
  parseIntegrationsOauthClientsJson,
5435
7084
  parseMcpServers,
@@ -5462,23 +7111,33 @@ export {
5462
7111
  resolveVoiceInputProviderRegistry,
5463
7112
  responseSatisfiesLatencyMode,
5464
7113
  retryStartupDependency,
7114
+ reviewedModelListPricing,
5465
7115
  runnableLatencyModesForModel,
5466
7116
  sandboxArchiveCaptureTimeoutMs,
5467
7117
  sandboxEnvironmentVariableNames,
7118
+ sandboxImageAllowlist,
5468
7119
  sandboxLifecycleHookIds,
5469
7120
  sandboxLifecycleTransitionWaitMs,
5470
7121
  sandboxPreparationProfiles,
5471
7122
  sandboxWarmRateMicrosPerSecond,
5472
7123
  selectModelPricing,
7124
+ selectSubscriptionAccount,
5473
7125
  serviceTierForLatencyMode,
7126
+ settingsForAcceptedSubscriptionTurn,
5474
7127
  settingsWithResolvedModelContext,
5475
7128
  stableSandboxEnvironmentForRun,
5476
7129
  startupRetryOptions,
5477
7130
  streamTokenDegraded,
7131
+ subscriptionAccountShardIndex,
5478
7132
  temporalConnectionOptions,
7133
+ trustedProxyCidrEntries,
7134
+ usableJevApiKey,
5479
7135
  validateModelCatalogSettings,
5480
7136
  voiceInputDeploymentConfigured,
7137
+ withClaudeConnectionCatalog,
7138
+ withClaudeConnectionCredential,
5481
7139
  withCodexCatalogProvider,
7140
+ withDirectModelProviders,
5482
7141
  withOrganizationGatewayCatalogProvider,
5483
7142
  withOrganizationGatewayCredential,
5484
7143
  withOrganizationOpenRouterCatalogProvider,