@bitkyc08/opencodex 2.40.0 → 2.41.0

package/src/server/responses/core.ts

@@ -68,12 +68,14 @@ import { resolvePassiveRouteSubjectId } from "../passive-route-linker";
 import {
   advanceComboAfterFailure,
   comboDefaultEffort,
+  comboFailureCooldownScope,
   comboFailureDecision,
   comboIdFromRawBody,
   comboRequestHasImageInput,
   concreteComboRequestBody,
   getCombo,
   isComboTargetInCooldown,
+  comboCooldownRetryAfterSeconds,
   NoAvailableComboTargetsError,
   noteComboSuccess,
   parseRetryAfterMs,
@@ -326,6 +328,7 @@ import {
   restoreImageGenCallsInJson,
 } from "../responses-image-gen-repair";
 import { createResponsesModelPayloadRewrite, rewriteResponsesModelJson } from "../responses-model-rewrite";
+import { parseRequestEffortRowId } from "../effort-row";
 import {
   collectSelfNamedNamespaceScrubAuthorization,
   createSelfNamedToolCallNamespaceScrubRewrite,
@@ -1434,15 +1437,29 @@ export function decodeRequestErrorResponse(err: unknown, label: string): Respons
 
 
 
-export function comboUnavailableResponse(message: string): Response {
+export function comboUnavailableResponse(
+  message: string,
+  options?: { retryAfter?: string | null },
+): Response {
+  const headers = new Headers({ "Content-Type": "application/json" });
+  const retryAfter = options?.retryAfter?.trim();
+  if (retryAfter && retryAfter.length > 0 && retryAfter.length <= 128) {
+    headers.set("Retry-After", retryAfter);
+  }
   return new Response(
     JSON.stringify({
       error: { message, type: "server_error", code: "combo_unavailable" },
     }),
-    { status: 503, headers: { "Content-Type": "application/json" } },
+    { status: 503, headers },
   );
 }
 
+function comboUnavailable(comboId: string, now = Date.now()): Response {
+  return comboUnavailableResponse(`No available targets for combo: ${comboId}`, {
+    retryAfter: comboCooldownRetryAfterSeconds(comboId, now),
+  });
+}
+
 
 
 export interface ConsumedComboFailure {
@@ -2297,7 +2314,7 @@ export async function handleComboResponses(
         config,
         { parentThreadId: inboundClientThreadId },
       );
-      return comboUnavailableResponse(`No available targets for combo: ${comboId}`);
+      return comboUnavailable(comboId);
     }
     let recovered = false;
     try {
@@ -2334,7 +2351,7 @@ export async function handleComboResponses(
   }
 
   if (!pick) {
-    return comboUnavailableResponse(`No available targets for combo: ${comboId}`);
+    return comboUnavailable(comboId);
   }
   // One immutable combo selection trace, before any child dispatch; child
   // adoption below must never replace it with a concrete child route trace.
@@ -2541,7 +2558,13 @@ export async function handleComboResponses(
     const nextPick = advanceComboAfterFailure(config, pick, {
       retryAfter: failure.retryAfter,
       now: Date.now(),
+      cooldownScope: comboFailureCooldownScope(failure.response.status, failure.classificationText, {
+        code: failure.upstreamCode,
+      }),
       eligible: payloadEligible,
+      status: failure.response.status,
+      code: failure.upstreamCode,
+      message: failure.classificationText,
     });
     if (!nextPick) adoptFailedChildLog(childLog);
     pick = nextPick;
@@ -2697,6 +2720,23 @@ async function handleResponsesInner(
     }
     return decodeRequestErrorResponse(err, "responses");
   }
+  // An effort row naming a table-less combo (`combo/x--high`) must reach the combo dispatcher
+  // as its base id, so the selector is normalized here, before comboIdFromRawBody reads model.
+  const comboEffortRow = !options.comboAttempt && body && typeof body === "object" && !Array.isArray(body)
+    && typeof (body as { model?: unknown }).model === "string"
+    ? parseRequestEffortRowId((body as { model: string }).model, config)
+    : null;
+  if (comboEffortRow) {
+    const raw = body as Record<string, unknown>;
+    raw.model = comboEffortRow.baseId;
+    const rawReasoning = raw.reasoning;
+    raw.reasoning = {
+      ...(rawReasoning && typeof rawReasoning === "object" && !Array.isArray(rawReasoning)
+        ? rawReasoning as Record<string, unknown>
+        : {}),
+      effort: comboEffortRow.effort,
+    };
+  }
   const comboId = !options.comboAttempt ? comboIdFromRawBody(body, config) : null;
   if (comboId && Object.hasOwn(config.combos ?? {}, comboId)) {
     options.onRequestBodyRead?.();
@@ -2751,6 +2791,20 @@ async function handleResponsesInner(
   let toolBridgeMaps: ReturnType<typeof buildToolBridgeMaps>;
   try {
     parsed = parseRequest(body);
+    const effortRow = parseRequestEffortRowId(parsed.modelId, config);
+    if (effortRow) {
+      parsed.modelId = effortRow.baseId;
+      parsed.options.reasoning = effortRow.effort;
+      const raw = parsed._rawBody as Record<string, unknown>;
+      const rawReasoning = raw.reasoning;
+      raw.model = effortRow.baseId;
+      raw.reasoning = {
+        ...(rawReasoning && typeof rawReasoning === "object" && !Array.isArray(rawReasoning)
+          ? rawReasoning as Record<string, unknown>
+          : {}),
+        effort: effortRow.effort,
+      };
+    }
     if (options.comboReplaySnapshot?.recoveredPlaintext) {
       markBodyNonPersistable(parsed._rawBody);
     }
@@ -2881,7 +2935,7 @@ async function handleResponsesInner(
     logCtx.routeDecision = route.routeDecision;
   } catch (err) {
     if (err instanceof NoAvailableComboTargetsError) {
-      return comboUnavailableResponse(err.message);
+      return comboUnavailable(err.comboId);
     }
     if (err instanceof NoEligiblePolicyCandidateError) {
       // Persist the evaluation trace (per-candidate exclusions + the
@@ -2991,7 +3045,7 @@ async function handleResponsesInner(
         logCtx.routeDecision = route.routeDecision;
       } catch (err) {
         if (err instanceof NoAvailableComboTargetsError) {
-          return comboUnavailableResponse(err.message);
+          return comboUnavailable(err.comboId);
         }
         if (err instanceof NoEligiblePolicyCandidateError) {
           logCtx.routeDecision = err.trace;
@@ -3115,7 +3169,7 @@ async function handleResponsesInner(
               logCtx.routeDecision = route.routeDecision;
             } catch (err) {
               if (err instanceof NoAvailableComboTargetsError) {
-                return comboUnavailableResponse(err.message);
+                return comboUnavailable(err.comboId);
               }
               if (err instanceof NoEligiblePolicyCandidateError) {
                 logCtx.routeDecision = err.trace;

package/src/types/config.ts

@@ -348,7 +348,10 @@ export interface OcxConfig {
    * the guess is wrong.
    */
   oauthOpenBrowser?: boolean;
-  /** Maximum usage-log bytes read for one management snapshot. */
+  /**
+   * @deprecated Compatibility-only limit for bounded legacy usage readers.
+   * `GET /api/usage` always aggregates the complete ledger.
+   */
   managementUsageMaxReadBytes?: number;
   providers: Record<string, OcxProviderConfig>;
   defaultProvider: string;
@@ -366,6 +369,12 @@ export interface OcxConfig {
   };
   /** Enable the shipped model alias patterns for providers without an override. */
   defaultModelAliases?: boolean;
+  /**
+   * Opt-in Cursor Private Inference compatibility rows. When true, `/v1/models`
+   * adds `<base-id>--<effort>` selectors for reasoning-capable model ids absent
+   * from Cursor's built-in effort table. Omitted/false preserves discovery output.
+   */
+  cursorEffortRows?: boolean;
   /** Explicit top-level deletion intent used by stale whole-config rebases. */
   configRebaseProvenance?: OcxConfigRebaseProvenance | Record<string, unknown>;
   /** OpenAI provider-contract migration marker (v2 = single `openai` provider with account mode). */

package/src/types/tools.ts

@@ -35,20 +35,23 @@ export function namespacedToolName(namespace: string | undefined, name: string):
  * Codex unified-exec name normalization.
  *
  * Codex's code-mode shell tool is declared as `exec` (a freeform custom tool whose own
- * description mentions the nested `await tools.exec_command(...)` helper). Routed models —
- * DeepSeek in particular — sometimes echo that helper name as the tool-call name, emitting
- * `exec_command` or `apply_patch` instead of the declared `exec`. Accept these nested helper
- * names only when the request catalog actually declares `exec` and does not itself declare the
- * emitted name (an MCP server may legitimately advertise one under its own namespace).
+ * description mentions the nested `await tools.exec_command(...)` helper). Some routed providers
+ * echo that helper name as the tool-call name, emitting `exec_command`, `write_stdin`, or
+ * `apply_patch` instead of the declared `exec`. Accept these nested helper names only when the
+ * request catalog actually declares `exec` and does not itself declare the emitted name (an MCP
+ * server may legitimately advertise one under its own namespace).
  */
 const LEGACY_SHELL_BRIDGE_TOOL_NAMES = ["exec_command", "shell_command"] as const;
-const CODE_MODE_HELPER_TOOL_NAMES = [...LEGACY_SHELL_BRIDGE_TOOL_NAMES, "apply_patch"] as const;
+const CODE_MODE_HELPER_TOOL_NAMES = [
+  ...LEGACY_SHELL_BRIDGE_TOOL_NAMES,
+  "write_stdin",
+  "apply_patch",
+] as const;
 
 /**
  * The one declared name that turns nested-helper normalization on. Declaring it is not just a
- * name: it also decides whether an emitted `exec_command`/`shell_command`/`apply_patch` is
- * accepted as that shell tool, so callers that build declared-name sets must add it only for a
- * genuine bare declaration.
+ * name: it also decides whether an emitted helper name is accepted as that shell tool, so callers
+ * that build declared-name sets must add it only for a genuine bare declaration.
  */
 export const CODE_MODE_EXEC_TOOL_NAME = "exec";
 

package/src/usage/expected-prices.ts

@@ -1,3 +1,5 @@
+import { normalizeCursorClaudeId } from "../adapters/cursor/claude-id";
+
 /**
  * Expected-price overlay for models whose jawcode cost rows are missing or all-zero
  * (subscription/OAuth surfaces). Sourced from official pricing pages only
@@ -56,6 +58,10 @@ const GEMINI_36_FLASH: Cost4 = { input: 1.5, output: 7.5, cacheRead: 0.15, cache
 // through 2026-12-31, stepping up to $1.50 / $7.50 on 2027-01-01. Revisit this row
 // then — the promotional rate is dated on the pricing page, not open-ended.
 const GEMINI_37_FLASH: Cost4 = { input: 0.75, output: 3.75, cacheRead: 0.075, cacheWrite: 0 };
+// Gemini 3.8 Flash carries the same published promotional shape as 3.7 through 2026-12-31,
+// rising to $1.50 / $7.50 on 2027-01-01. A SEPARATE constant on purpose: equal today, but
+// aliasing them would silently drag 3.8 along if 3.7's row is ever re-verified differently.
+const GEMINI_38_FLASH: Cost4 = { input: 0.75, output: 3.75, cacheRead: 0.075, cacheWrite: 0 };
 const MINIMAX_M21_HIGHSPEED: Cost4 = { input: 0.6, output: 2.4, cacheRead: 0.03, cacheWrite: 0.375 };
 const KIMI_K3: Cost4 = { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3 };
 const KIMI_K27_CODE: Cost4 = { input: 0.95, output: 4, cacheRead: 0.19, cacheWrite: 0.95 };
@@ -79,8 +85,19 @@ const ANTHROPIC_PRICING = "https://platform.claude.com/docs/en/about-claude/pric
 
 const GEMINI_PRICING = "https://ai.google.dev/gemini-api/docs/pricing (2026-07-22); cacheWrite=0: storage is billed per-hour, not per-token";
 const GEMINI_37_PRICING = "https://ai.google.dev/gemini-api/docs/pricing (2026-08-14); promotional rate through 2026-12-31, rises to 1.50/7.50 on 2027-01-01; cacheWrite=0: storage is billed per-hour, not per-token";
+const GEMINI_38_PRICING = "https://ai.google.dev/gemini-api/docs/pricing (2026-09-03); promotional rate through 2026-12-31, rises to 1.50/7.50 on 2027-01-01; cacheWrite=0: storage is billed per-hour, not per-token";
 const MINIMAX_PRICING = "https://platform.minimax.io/docs/guides/pricing-paygo";
 const OPENAI_GPT56_PRICING = "https://developers.openai.com/api/docs/pricing";
+const META_MODEL_PRICING = "https://dev.meta.ai/docs/pricing-rate-limits";
+/*
+ * Shared by both Meta providers. Overlays resolve by EXACT provider id, so `meta-muse`
+ * cannot inherit `meta-model`'s rows — and an unpriced provider whose whole warning is
+ * "treat every call as billable" would report no cost at all.
+ */
+const META_MUSE_SPARK_13: Cost4 = { input: 1.25, output: 4.25, cacheRead: 0.15, cacheWrite: 0 };
+const META_MUSE_SPARK_13_CONTRIBUTOR: Cost4 = { input: 0.1, output: 0.2, cacheRead: 0.002, cacheWrite: 0 };
+const META_SPARK_SOURCE = `Meta Model API published price ${META_MODEL_PRICING}`;
+const META_SPARK_CONTRIBUTOR_SOURCE = `Meta Model API published Contributor-tier price ${META_MODEL_PRICING}; data-sharing discount tier`;
 const DEEPSEEK_PRICING = "https://api-docs.deepseek.com/quick_start/pricing-details-usd; V4 Flash alias transition scheduled 2026-07-24 — re-verify after";
 // Kimi official tables publish input/output/cache-hit only; cacheWrite is mapped to the
 // cache-miss input price (Kimi auto-caches with no separate write billing). 2026-07-20 re-verified.
@@ -100,13 +117,8 @@ export const EXPECTED_PRICE_OVERLAYS: readonly ExpectedPriceOverlay[] = [
   // bundle collapses anthropic-apikey onto anthropic).
   { provider: "anthropic", modelId: "claude-fable-5-1", cost4: CLAUDE_FABLE_51, source: `anthropic official Claude Fable 5.1 ${ANTHROPIC_PRICING}; cache hit = 0.025x base input`, verifiedAt: "2026-09-02", status: "verified" },
   { provider: "anthropic-apikey", modelId: "claude-fable-5-1", cost4: CLAUDE_FABLE_51, source: `anthropic official Claude Fable 5.1 ${ANTHROPIC_PRICING}; cache hit = 0.025x base input`, verifiedAt: "2026-09-02", status: "verified" },
-  // Cursor seeds Fable 5.1 preemptively under three spellings (adapters/cursor/catalog.ts);
-  // the model-level vendor fallback only searches jawcode metadata, which has no Fable 5.1
-  // row yet, so each Cursor spelling needs its own overlay. Vendor list price, like the
-  // cursor/claude-opus-5 row.
+  // Cursor canonicalizes every Fable 5.1 spelling onto this sole overlay row.
   { provider: "cursor", modelId: "claude-fable-5-1", cost4: CLAUDE_FABLE_51, source: `anthropic official Claude Fable 5.1 ${ANTHROPIC_PRICING}; cache hit = 0.025x base input; vendor list price applied to the Cursor surface`, verifiedAt: "2026-09-02", status: "verified-derived" },
-  { provider: "cursor", modelId: "claude-fable-5.1", cost4: CLAUDE_FABLE_51, source: `anthropic official Claude Fable 5.1 ${ANTHROPIC_PRICING}; cache hit = 0.025x base input; vendor list price applied to the Cursor surface`, verifiedAt: "2026-09-02", status: "verified-derived" },
-  { provider: "cursor", modelId: "claude-5.1-fable", cost4: CLAUDE_FABLE_51, source: `anthropic official Claude Fable 5.1 ${ANTHROPIC_PRICING}; cache hit = 0.025x base input; vendor list price applied to the Cursor surface`, verifiedAt: "2026-09-02", status: "verified-derived" },
   // claude-opus-5 is exposed by three providers but absent from the jawcode bundle, so
   // cost resolution returned null and the Logs `~$` column rendered an em dash. The
   // model-level vendor fallback only searches jawcode metadata, never overlays, so one
@@ -126,6 +138,10 @@ export const EXPECTED_PRICE_OVERLAYS: readonly ExpectedPriceOverlay[] = [
   // 3.7 Flash rides CCA, whose billing equivalence to the Developer API list price is
   // not published, so this is `verified-derived` rather than `verified`: the number is
   // proven, the claim that Antigravity charges it is inferred.
+  { provider: "google-antigravity", modelId: "gemini-3.8-flash", cost4: GEMINI_38_FLASH, source: `derived: Gemini 3.8 Flash promotional rate through 2026-12-31 ${GEMINI_38_PRICING}`, verifiedAt: "2026-09-03", status: "verified-derived" },
+  { provider: "google-antigravity", modelId: "gemini-3.8-flash-low", cost4: GEMINI_38_FLASH, source: `derived: gemini-3.8-flash ${GEMINI_38_PRICING}`, verifiedAt: "2026-09-03", status: "verified-derived" },
+  { provider: "google-antigravity", modelId: "gemini-3.8-flash-medium", cost4: GEMINI_38_FLASH, source: `derived: gemini-3.8-flash ${GEMINI_38_PRICING}`, verifiedAt: "2026-09-03", status: "verified-derived" },
+  { provider: "google-antigravity", modelId: "gemini-3.8-flash-high", cost4: GEMINI_38_FLASH, source: `derived: gemini-3.8-flash ${GEMINI_38_PRICING}`, verifiedAt: "2026-09-03", status: "verified-derived" },
   { provider: "google-antigravity", modelId: "gemini-3.7-flash", cost4: GEMINI_37_FLASH, source: `derived: Gemini 3.7 Flash promotional rate through 2026-12-31 ${GEMINI_37_PRICING}`, verifiedAt: "2026-08-14", status: "verified-derived" },
   // Retained after the 3.6 retirement: historical usage.jsonl rows still carry these
   // ids, and dropping the row would silently zero the cost of requests already made.
@@ -139,6 +155,19 @@ export const EXPECTED_PRICE_OVERLAYS: readonly ExpectedPriceOverlay[] = [
   { provider: "openai-apikey", modelId: "gpt-5.6-sol-pro", cost4: GPT56_SOL, source: `collapsed base ID ${OPENAI_GPT56_PRICING}`, verifiedAt: "2026-08-03", status: "verified-derived" },
   { provider: "openai-apikey", modelId: "gpt-5.6-terra-pro", cost4: GPT56_TERRA, source: `collapsed base ID ${OPENAI_GPT56_PRICING}`, verifiedAt: "2026-08-03", status: "verified-derived" },
   { provider: "openai-apikey", modelId: "gpt-5.6-luna-pro", cost4: GPT56_LUNA, source: `collapsed base ID ${OPENAI_GPT56_PRICING}`, verifiedAt: "2026-08-03", status: "verified-derived" },
+  // Meta Model API direct provider. `meta-model` has no jawcode metadata alias, so an
+  // unpriced row falls through the whole resolution chain and the Logs cost column
+  // renders nothing — these exact overlays are the only source. Both are Meta's own
+  // published list prices for Meta's own endpoint (hence "verified", not derived), and
+  // they match the figures Command Code republishes for the same two models.
+  // cacheWrite=0: Meta publishes a cached-input price but no cache-write charge.
+  { provider: "meta-model", modelId: "muse-spark-1.3", cost4: META_MUSE_SPARK_13, source: META_SPARK_SOURCE, verifiedAt: "2026-09-03", status: "verified" },
+  { provider: "meta-model", modelId: "muse-spark-1.3-contributor", cost4: META_MUSE_SPARK_13_CONTRIBUTOR, source: META_SPARK_CONTRIBUTOR_SOURCE, verifiedAt: "2026-09-03", status: "verified" },
+  // Same endpoint, same list price, different credential. Meta does not authorize this
+  // reuse and settlement is not observable, so these are the public Model API rates as a
+  // conservative estimate — not evidence of how the call is actually billed.
+  { provider: "meta-muse", modelId: "muse-spark-1.3", cost4: META_MUSE_SPARK_13, source: META_SPARK_SOURCE, verifiedAt: "2026-09-03", status: "verified-derived" },
+  { provider: "meta-muse", modelId: "muse-spark-1.3-contributor", cost4: META_MUSE_SPARK_13_CONTRIBUTOR, source: META_SPARK_CONTRIBUTOR_SOURCE, verifiedAt: "2026-09-03", status: "verified-derived" },
   // Daybreak aliases: priced as their current snapshots (red -> gpt-5.6-cyber,
   // blue -> gpt-5.6-sol). The alias ids carry no rows of their own upstream, hence
   // verified-derived. Blue deliberately reuses GPT56_SOL rather than duplicating the tuple.
@@ -163,6 +192,7 @@ export const EXPECTED_PRICE_OVERLAYS: readonly ExpectedPriceOverlay[] = [
   { provider: "google", modelId: "gemini-3.6-flash", cost4: GEMINI_36_FLASH, source: GEMINI_PRICING, verifiedAt: "2026-07-22", status: "verified" },
   // Developer API row: the price IS published for this surface, so `verified`.
   { provider: "google", modelId: "gemini-3.7-flash", cost4: GEMINI_37_FLASH, source: GEMINI_37_PRICING, verifiedAt: "2026-08-14", status: "verified" },
+  { provider: "google", modelId: "gemini-3.8-flash", cost4: GEMINI_38_FLASH, source: GEMINI_38_PRICING, verifiedAt: "2026-09-03", status: "verified" },
   { provider: "google-antigravity", modelId: "gemini-3.1-pro-preview", cost4: GEMINI_31_PRO, source: GEMINI_PRICING, verifiedAt: "2026-07-20", status: "verified" },
   // Antigravity-bundled third-party models — derived from the underlying vendor's
   // official API price (Antigravity itself bills via subscription quota).
@@ -236,8 +266,14 @@ export function findExpectedPriceOverlay(
   overlays: readonly ExpectedPriceOverlay[] = EXPECTED_PRICE_OVERLAYS,
 ): ExpectedPriceOverlay | undefined {
   const exact = overlays.filter(row => row.provider === provider && row.modelId === modelId);
-  return exact.find(row => row.status === "verified")
+  const match = exact.find(row => row.status === "verified")
     ?? exact.find(row => row.status === "verified-derived");
+  if (match || provider !== "cursor") return match;
+  const canonicalBaseId = normalizeCursorClaudeId(modelId)?.canonicalBaseId;
+  if (!canonicalBaseId) return undefined;
+  const canonical = overlays.filter(row => row.provider === provider && row.modelId === canonicalBaseId);
+  return canonical.find(row => row.status === "verified")
+    ?? canonical.find(row => row.status === "verified-derived");
 }
 
 /** OpenAI Fast price multipliers retained as a compatibility export. */