@lightsage/mcp 0.2.2 → 0.3.1

package/src/models/v2evalrunsanalysisop.ts

@@ -0,0 +1,48 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: 4c00eef582d7
+ */
+
+import * as z from "zod";
+import {
+  HTTPValidationError,
+  HTTPValidationError$zodSchema,
+} from "./httpvalidationerror.js";
+import {
+  V2EvalRunAnalysisResponse,
+  V2EvalRunAnalysisResponse$zodSchema,
+} from "./v2evalrunanalysisresponse.js";
+
+export type V2EvalRunsAnalysisRequest = { run_id: string };
+
+export const V2EvalRunsAnalysisRequest$zodSchema: z.ZodType<
+  V2EvalRunsAnalysisRequest
+> = z.object({
+  run_id: z.string(),
+});
+
+/**
+ * Eval run not found. Use the run ID returned when you started the eval.
+ */
+export type V2EvalRunsAnalysisResponseBody = { detail: string };
+
+export const V2EvalRunsAnalysisResponseBody$zodSchema: z.ZodType<
+  V2EvalRunsAnalysisResponseBody
+> = z.object({
+  detail: z.string(),
+}).describe(
+  "Eval run not found. Use the run ID returned when you started the eval.",
+);
+
+export type V2EvalRunsAnalysisResponse =
+  | V2EvalRunAnalysisResponse
+  | V2EvalRunsAnalysisResponseBody
+  | HTTPValidationError;
+
+export const V2EvalRunsAnalysisResponse$zodSchema: z.ZodType<
+  V2EvalRunsAnalysisResponse
+> = z.union([
+  V2EvalRunAnalysisResponse$zodSchema,
+  z.lazy(() => V2EvalRunsAnalysisResponseBody$zodSchema),
+  HTTPValidationError$zodSchema,
+]);

package/src/models/v2evalrunstraceop.ts

@@ -0,0 +1,58 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: f7cdcd3e7f8d
+ */
+
+import * as z from "zod";
+import {
+  HTTPValidationError,
+  HTTPValidationError$zodSchema,
+} from "./httpvalidationerror.js";
+import {
+  V2EvalRunTraceResponse,
+  V2EvalRunTraceResponse$zodSchema,
+} from "./v2evalruntraceresponse.js";
+
+export type V2EvalRunsTraceRequest = {
+  run_id: string;
+  after_id?: number | undefined;
+  limit?: number | undefined;
+};
+
+export const V2EvalRunsTraceRequest$zodSchema: z.ZodType<
+  V2EvalRunsTraceRequest
+> = z.object({
+  after_id: z.int().default(0).describe(
+    "Return events with IDs greater than this cursor.",
+  ),
+  limit: z.int().default(500).describe(
+    "Maximum number of trace events to return.",
+  ),
+  run_id: z.string(),
+});
+
+/**
+ * Eval run not found. Use the run ID returned when you started the eval.
+ */
+export type V2EvalRunsTraceResponseBody = { detail: string };
+
+export const V2EvalRunsTraceResponseBody$zodSchema: z.ZodType<
+  V2EvalRunsTraceResponseBody
+> = z.object({
+  detail: z.string(),
+}).describe(
+  "Eval run not found. Use the run ID returned when you started the eval.",
+);
+
+export type V2EvalRunsTraceResponse =
+  | V2EvalRunTraceResponse
+  | V2EvalRunsTraceResponseBody
+  | HTTPValidationError;
+
+export const V2EvalRunsTraceResponse$zodSchema: z.ZodType<
+  V2EvalRunsTraceResponse
+> = z.union([
+  V2EvalRunTraceResponse$zodSchema,
+  z.lazy(() => V2EvalRunsTraceResponseBody$zodSchema),
+  HTTPValidationError$zodSchema,
+]);

package/src/models/v2evalruntraceevent.ts

@@ -0,0 +1,45 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: cf33bec99838
+ */
+
+import * as z from "zod";
+
+export type V2EvalRunTraceEvent = {
+  run_id: string;
+  eval_id?: string | null | undefined;
+  eval_name?: string | null | undefined;
+  configuration_id?: string | null | undefined;
+  agent?: string | null | undefined;
+  prompt_index?: number | undefined;
+  repeat_index?: number | undefined;
+  child_job_id: string;
+  event_id: number;
+  timestamp?: string | null | undefined;
+  category: string;
+  type: string;
+  message: string;
+  data?: { [k: string]: any } | undefined;
+};
+
+export const V2EvalRunTraceEvent$zodSchema: z.ZodType<V2EvalRunTraceEvent> = z
+  .object({
+    agent: z.string().nullable().optional(),
+    category: z.string(),
+    child_job_id: z.string().describe(
+      "Child eval-attempt ID associated with the event.",
+    ),
+    configuration_id: z.string().nullable().optional(),
+    data: z.record(z.string(), z.any()).optional().describe(
+      "Sanitized event payload.",
+    ),
+    eval_id: z.string().nullable().optional(),
+    eval_name: z.string().nullable().optional(),
+    event_id: z.int(),
+    message: z.string(),
+    prompt_index: z.int().default(0),
+    repeat_index: z.int().default(0),
+    run_id: z.string(),
+    timestamp: z.string().nullable().optional(),
+    type: z.string(),
+  });

package/src/models/v2evalruntraceresponse.ts

@@ -0,0 +1,38 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: 25d2b266ce5f
+ */
+
+import * as z from "zod";
+import {
+  PublicEvalRunProgress,
+  PublicEvalRunProgress$zodSchema,
+} from "./publicevalrunprogress.js";
+import {
+  V2EvalRunTraceEvent,
+  V2EvalRunTraceEvent$zodSchema,
+} from "./v2evalruntraceevent.js";
+
+export type V2EvalRunTraceResponse = {
+  run_id: string;
+  status: string;
+  progress?: PublicEvalRunProgress | undefined;
+  error?: string | null | undefined;
+  data?: Array<V2EvalRunTraceEvent> | undefined;
+  next_after_id: number;
+  has_more: boolean;
+};
+
+export const V2EvalRunTraceResponse$zodSchema: z.ZodType<
+  V2EvalRunTraceResponse
+> = z.object({
+  data: z.array(V2EvalRunTraceEvent$zodSchema).optional(),
+  error: z.string().nullable().optional(),
+  has_more: z.boolean(),
+  next_after_id: z.int().describe(
+    "Cursor to pass as `after_id` when requesting the next page.",
+  ),
+  progress: PublicEvalRunProgress$zodSchema.optional(),
+  run_id: z.string(),
+  status: z.string(),
+});

package/src/models/v2inlineclireference.ts

@@ -0,0 +1,15 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: d13cb884e6f6
+ */
+
+import * as z from "zod";
+
+export type V2InlineCliReference = { install_command: string };
+
+export const V2InlineCliReference$zodSchema: z.ZodType<V2InlineCliReference> = z
+  .object({
+    install_command: z.string().describe(
+      "Shell command to install this CLI before the agent starts.",
+    ),
+  });

package/src/models/v2inlinemcpdefinition.ts

@@ -0,0 +1,51 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: 96ca65e675c2
+ */
+
+import * as z from "zod";
+import { ClosedEnum } from "../types/enums.js";
+
+export const Type = {
+  Http: "http",
+  Sse: "sse",
+  Stdio: "stdio",
+} as const;
+export type Type = ClosedEnum<typeof Type>;
+
+export const Type$zodSchema = z.enum([
+  "http",
+  "sse",
+  "stdio",
+]);
+
+export type V2InlineMcpDefinition = {
+  name: string;
+  type: Type;
+  url?: string | null | undefined;
+  headers?: { [k: string]: string } | undefined;
+  command?: string | null | undefined;
+  args?: Array<string> | undefined;
+  env?: { [k: string]: string } | undefined;
+};
+
+export const V2InlineMcpDefinition$zodSchema: z.ZodType<V2InlineMcpDefinition> =
+  z.object({
+    args: z.array(z.string()).optional(),
+    command: z.string().nullable().optional().describe(
+      "Executable command for a `stdio` MCP server.",
+    ),
+    env: z.record(z.string(), z.string()).optional().describe(
+      "Environment mapping for a `stdio` MCP server. Values may reference run secrets with `${VARIABLE_NAME}`.",
+    ),
+    headers: z.record(z.string(), z.string()).optional().describe(
+      "HTTP headers. Reference run secrets with `${VARIABLE_NAME}` instead of embedding credentials.",
+    ),
+    name: z.string().describe(
+      "Stable server name used in the selected agent's MCP config.",
+    ),
+    type: Type$zodSchema,
+    url: z.string().nullable().optional().describe(
+      "HTTPS endpoint for an `http` or `sse` MCP server.",
+    ),
+  });

package/src/models/v2savedclireference.ts

@@ -0,0 +1,13 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: 99afd03df958
+ */
+
+import * as z from "zod";
+
+export type V2SavedCliReference = { id: string };
+
+export const V2SavedCliReference$zodSchema: z.ZodType<V2SavedCliReference> = z
+  .object({
+    id: z.string().describe("Saved CLI ID from `GET /v2/clis`."),
+  });

package/src/models/v2savedmcpreference.ts

@@ -0,0 +1,13 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: 9cc0c2e77b20
+ */
+
+import * as z from "zod";
+
+export type V2SavedMcpReference = { id: string };
+
+export const V2SavedMcpReference$zodSchema: z.ZodType<V2SavedMcpReference> = z
+  .object({
+    id: z.string().describe("Saved MCP server ID from `GET /v2/mcp-servers`."),
+  });

package/src/tool-names.ts

@@ -110,19 +110,19 @@ export const toolNames: Array<{ name: string; description: string }>= [
   },
   {
     "name": "evals-v2-evals-list",
-    "description": "List Evals\n\nReturns all active evals including workflow, MCP, CLI, and skill evals owned by the organization."
+    "description": "List Evals\n\nReturns all active eval definitions owned by the organization."
   },
   {
     "name": "evals-v2-evals-create",
-    "description": "Create Eval\n\nCreates an eval with a task prompt, subject, and judge. The eval can then be run through eval runs."
+    "description": "Create Eval\n\nCreates a reusable simulation definition with one or more task prompts and judges. Optional agents, persona, CLI installations, MCP servers, repeat count, and tags are saved with the eval. Creating the eval does not start a run and does not accept secret environment values."
   },
   {
     "name": "evals-v2-evals-retrieve",
-    "description": "Retrieve Eval\n\nReturns one eval and its judge configuration."
+    "description": "Retrieve Eval\n\nReturns one reusable eval simulation definition."
   },
   {
     "name": "evals-v2-evals-update",
-    "description": "Update Eval\n\nUpdates an eval's task prompt, subject, judge, or name. Omitted fields remain unchanged."
+    "description": "Update Eval\n\nUpdates a reusable eval simulation definition. Omitted fields remain unchanged. Empty `agent`, `clis`, `mcps`, or `tags` arrays clear those collections, and `persona: null` clears the saved persona."
   },
   {
     "name": "evals-v2-evals-delete",
@@ -150,11 +150,15 @@ export const toolNames: Array<{ name: string; description: string }>= [
   },
   {
     "name": "evals-run",
-    "description": "Start an eval run\n\nQueues a durable eval job and returns immediately with a polling ID. Poll `GET /v2/eval-runs/{run_id}` until its status reaches a terminal value.\n\nSend one of these flat request forms:\n\n- **Saved eval:** provide `eval_id`, `configuration_id`, and `agent_id`.\n- **Direct create-and-run:** provide `prompt`, `judges`, and `agent_id`. You may also provide a repository, write-only environment variables, and public skill packages.\n\nDo not combine saved-run and direct-run fields. No separate `mode` field is required."
+    "description": "Start an eval run\n\nQueues a durable eval job and returns immediately with a polling ID. Poll `GET /v2/eval-runs/{run_id}` until its status reaches a terminal value.\n\nSend one of these flat request forms:\n\n- **Saved eval:** provide `eval_id`; optionally replace its prompts, persona, agents, CLIs, MCPs, or runs for this launch. A saved configuration can still provide repository, environment, skills, and legacy tool selections.\n- **Direct create-and-run:** provide `prompts`, `judges`, and `agent`. You may also provide a persona, tags, repository, write-only environment variables, saved or inline CLIs and MCP servers, and skills.\n\nThe run fans out across prompts × agents × runs. Persona is shared metadata and does not multiply attempts. No separate `mode` field is required."
   },
   {
-    "name": "evals-monitor",
-    "description": "Monitor Runs\n\nStreams an immediate snapshot followed by live status and sanitized activity. Provide `run_id` to monitor one public v2 eval run, or omit it to capture all active runs when the connection opens."
+    "name": "evals-analysis",
+    "description": "Retrieve eval run analysis\n\nReturns the persisted analysis produced for each completed attempt in one eval run. The analysis includes the verdict, outcome explanation, findings, bottleneck, and analysis sections. Active runs return the analysis available so far together with current run status and progress."
+  },
+  {
+    "name": "evals-trace",
+    "description": "Retrieve an eval run trace\n\nReturns a finite page of sanitized execution events for one eval run. Use `next_after_id` as `after_id` to retrieve the next page. Use the monitor operation instead when following a run live."
   },
   {
     "name": "eval-runs-v2-eval-runs-list",
@@ -166,7 +170,7 @@ export const toolNames: Array<{ name: string; description: string }>= [
   },
   {
     "name": "eval-results-v2-eval-results-list",
-    "description": "List Results\n\nReturns completed custom-eval results from onboarding, v1, and v2 run flows, ordered newest first. Result IDs are different from eval run IDs."
+    "description": "List Results\n\nReturns completed custom-eval results from all supported run flows, ordered newest first. Result IDs are different from eval run IDs."
   },
   {
     "name": "eval-results-v2-eval-results-retrieve",

package/test/simulations.test.js

@@ -0,0 +1,162 @@
+import assert from "node:assert/strict";
+import http from "node:http";
+import test from "node:test";
+
+const { LightsageMcpCore } = await import("../esm/core.js");
+const { CreateV2EvalRequest$zodSchema } = await import(
+  "../esm/models/createv2evalrequest.js"
+);
+const { CreateV2EvalRunRequest$zodSchema } = await import(
+  "../esm/models/createv2evalrunrequest.js"
+);
+const { tool$evalsV2EvalsCreate } = await import(
+  "../esm/mcp-server/tools/evalsV2EvalsCreate.js"
+);
+const { tool$evalsRun } = await import(
+  "../esm/mcp-server/tools/evalsRun.js"
+);
+
+const savedCli = { id: "230f0c71-ed91-4a4a-9932-9cef267411bc" };
+const inlineCli = { install_command: "brew install lightsagehq/tools/lightsage" };
+const savedMcp = { id: "550e8400-e29b-41d4-a716-446655440000" };
+const inlineMcp = {
+  name: "lightsage",
+  type: "http",
+  url: "https://mcp.lightsage.com/mcp",
+  headers: {
+    "X-Lightsage-Api-Key": "${LIGHTSAGE_API_KEY}",
+  },
+};
+
+function createSimulationServer(requests) {
+  return http.createServer(async (req, res) => {
+    const chunks = [];
+    for await (const chunk of req) chunks.push(chunk);
+    requests.push({
+      method: req.method,
+      url: req.url,
+      apiKey: req.headers["x-lightsage-api-key"],
+      body: JSON.parse(Buffer.concat(chunks).toString()),
+    });
+    res.setHeader("content-type", "application/json");
+    res.end(JSON.stringify({ accepted: true }));
+  });
+}
+
+test("create eval accepts the canonical simulation matrix", () => {
+  const parsed = CreateV2EvalRequest$zodSchema.parse({
+    name: "Lightsage interface simulation",
+    prompts: ["Use the CLI.", "Use the MCP server."],
+    judges: ["The workflow succeeds."],
+    persona: "Platform engineer",
+    agent: ["codex:gpt-5-4", "claude-code:claude-opus-4-8"],
+    clis: [savedCli, inlineCli],
+    mcps: [savedMcp, inlineMcp],
+    runs: 2,
+    tags: ["smoke-test"],
+  });
+
+  assert.deepEqual(parsed.clis, [savedCli, inlineCli]);
+  assert.deepEqual(parsed.mcps, [savedMcp, inlineMcp]);
+  assert.equal(parsed.runs, 2);
+});
+
+test("direct eval run accepts the same canonical simulation fields", () => {
+  const parsed = CreateV2EvalRunRequest$zodSchema.parse({
+    prompts: ["Use the CLI.", "Use the MCP server."],
+    judges: ["The workflow succeeds."],
+    agent: ["codex:gpt-5-4", "claude-code:claude-opus-4-8"],
+    persona: "Platform engineer",
+    clis: [savedCli, inlineCli],
+    mcps: [savedMcp, inlineMcp],
+    env: { LIGHTSAGE_API_KEY: "runtime-only" },
+    skills: ["resend/resend-skills"],
+    runs: 3,
+  });
+
+  assert.deepEqual(parsed.agent, [
+    "codex:gpt-5-4",
+    "claude-code:claude-opus-4-8",
+  ]);
+  assert.deepEqual(parsed.clis, [savedCli, inlineCli]);
+  assert.deepEqual(parsed.mcps, [savedMcp, inlineMcp]);
+  assert.deepEqual(parsed.skills, ["resend/resend-skills"]);
+  assert.equal(parsed.runs, 3);
+});
+
+test("simulation tools reject non-object CLI and MCP entries", () => {
+  const result = CreateV2EvalRunRequest$zodSchema.safeParse({
+    prompts: ["Use the CLI."],
+    judges: ["The workflow succeeds."],
+    agent: ["codex:gpt-5-4"],
+    clis: ["brew install lightsage"],
+    mcps: ["https://mcp.lightsage.com/mcp"],
+  });
+
+  assert.equal(result.success, false);
+});
+
+test("MCP tools forward canonical eval resources unchanged", async () => {
+  const requests = [];
+  const server = createSimulationServer(requests);
+  await new Promise((resolve) => server.listen(0, "127.0.0.1", resolve));
+  const address = server.address();
+  const client = new LightsageMcpCore({
+    serverURL: `http://127.0.0.1:${address.port}`,
+    security: { LightsageApiKeyAuth: "ls_simulation_test" },
+  });
+  const ctx = { signal: new AbortController().signal };
+  const createRequest = {
+    name: "Lightsage interface simulation",
+    prompts: ["Use the CLI.", "Use the MCP server."],
+    judges: ["The workflow succeeds."],
+    agent: ["codex:gpt-5-4"],
+    clis: [savedCli, inlineCli],
+    mcps: [savedMcp, inlineMcp],
+    runs: 2,
+    tags: ["smoke-test"],
+  };
+  const runRequest = {
+    prompts: ["Use the CLI.", "Use the MCP server."],
+    judges: ["The workflow succeeds."],
+    agent: ["codex:gpt-5-4"],
+    clis: [savedCli, inlineCli],
+    mcps: [savedMcp, inlineMcp],
+    env: { LIGHTSAGE_API_KEY: "runtime-only" },
+    skills: ["resend/resend-skills"],
+    runs: 3,
+  };
+
+  try {
+    const createResult = await tool$evalsV2EvalsCreate.tool(
+      client,
+      { request: createRequest },
+      ctx,
+    );
+    assert.equal(createResult.isError, undefined);
+
+    const runResult = await tool$evalsRun.tool(
+      client,
+      { request: runRequest },
+      ctx,
+    );
+    assert.equal(runResult.isError, undefined);
+
+    assert.deepEqual(requests, [
+      {
+        method: "POST",
+        url: "/v2/evals",
+        apiKey: "ls_simulation_test",
+        body: createRequest,
+      },
+      {
+        method: "POST",
+        url: "/v2/eval-runs",
+        apiKey: "ls_simulation_test",
+        body: runRequest,
+      },
+    ]);
+  } finally {
+    server.close();
+  }
+});

package/esm/funcs/evalsMonitor.d.ts

@@ -1,20 +0,0 @@
-import { LightsageMcpCore } from "../core.js";
-import { RequestOptions } from "../lib/sdks.js";
-import { APIError } from "../models/errors/apierror.js";
-import { ConnectionError, InvalidRequestError, RequestAbortedError, RequestTimeoutError, UnexpectedClientError } from "../models/errors/httpclienterrors.js";
-import { SDKValidationError } from "../models/errors/sdkvalidationerror.js";
-import { V2EvalRunsMonitorRequest } from "../models/v2evalrunsmonitorop.js";
-import { APIPromise } from "../types/async.js";
-import { Result } from "../types/fp.js";
-export declare enum MonitorAcceptEnum {
-    applicationJsonAccept = "application/json",
-    textEventStreamAccept = "text/event-stream"
-}
-/**
- * Monitor Runs
- *
- * @remarks
- * Streams an immediate snapshot followed by live status and sanitized activity. Provide `run_id` to monitor one public v2 eval run, or omit it to capture all active runs when the connection opens.
- */
-export declare function evalsMonitor(client$: LightsageMcpCore, request?: V2EvalRunsMonitorRequest | undefined, options?: RequestOptions): APIPromise<Result<Response, APIError | SDKValidationError | UnexpectedClientError | InvalidRequestError | RequestAbortedError | RequestTimeoutError | ConnectionError>>;
-//# sourceMappingURL=evalsMonitor.d.ts.map

package/esm/funcs/evalsMonitor.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"evalsMonitor.d.ts","sourceRoot":"","sources":["../../src/funcs/evalsMonitor.ts"],"names":[],"mappings":"AAKA,OAAO,EAAE,gBAAgB,EAAE,MAAM,YAAY,CAAC;AAI9C,OAAO,EAAE,cAAc,EAAE,MAAM,gBAAgB,CAAC;AAGhD,OAAO,EAAE,QAAQ,EAAE,MAAM,8BAA8B,CAAC;AACxD,OAAO,EACL,eAAe,EACf,mBAAmB,EACnB,mBAAmB,EACnB,mBAAmB,EACnB,qBAAqB,EACtB,MAAM,sCAAsC,CAAC;AAC9C,OAAO,EAAE,kBAAkB,EAAE,MAAM,wCAAwC,CAAC;AAC5E,OAAO,EACL,wBAAwB,EAEzB,MAAM,kCAAkC,CAAC;AAC1C,OAAO,EAAW,UAAU,EAAE,MAAM,mBAAmB,CAAC;AACxD,OAAO,EAAE,MAAM,EAAE,MAAM,gBAAgB,CAAC;AAExC,oBAAY,iBAAiB;IAC3B,qBAAqB,qBAAqB;IAC1C,qBAAqB,sBAAsB;CAC5C;AAED;;;;;GAKG;AACH,wBAAgB,YAAY,CAC1B,OAAO,EAAE,gBAAgB,EACzB,OAAO,CAAC,EAAE,wBAAwB,GAAG,SAAS,EAC9C,OAAO,CAAC,EAAE,cAAc,GACvB,UAAU,CACX,MAAM,CACJ,QAAQ,EACN,QAAQ,GACR,kBAAkB,GAClB,qBAAqB,GACrB,mBAAmB,GACnB,mBAAmB,GACnB,mBAAmB,GACnB,eAAe,CAClB,CACF,CAMA"}

package/esm/funcs/evalsMonitor.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"evalsMonitor.js","sourceRoot":"","sources":["../../src/funcs/evalsMonitor.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAGH,OAAO,EAAE,eAAe,EAAE,MAAM,qBAAqB,CAAC;AACtD,OAAO,EAAE,UAAU,EAAE,MAAM,sBAAsB,CAAC;AAClD,OAAO,EAAE,SAAS,EAAE,MAAM,mBAAmB,CAAC;AAE9C,OAAO,EAAE,eAAe,EAAE,qBAAqB,EAAE,MAAM,oBAAoB,CAAC;AAC5E,OAAO,EAAE,UAAU,EAAE,MAAM,eAAe,CAAC;AAU3C,OAAO,EAEL,kCAAkC,GACnC,MAAM,kCAAkC,CAAC;AAC1C,OAAO,EAAW,UAAU,EAAE,MAAM,mBAAmB,CAAC;AAGxD,MAAM,CAAN,IAAY,iBAGX;AAHD,WAAY,iBAAiB;IAC3B,+DAA0C,CAAA;IAC1C,gEAA2C,CAAA;AAC7C,CAAC,EAHW,iBAAiB,KAAjB,iBAAiB,QAG5B;AAED;;;;;GAKG;AACH,MAAM,UAAU,YAAY,CAC1B,OAAyB,EACzB,OAA8C,EAC9C,OAAwB;IAaxB,OAAO,IAAI,UAAU,CAAC,GAAG,CACvB,OAAO,EACP,OAAO,EACP,OAAO,CACR,CAAC,CAAC;AACL,CAAC;AAED,KAAK,UAAU,GAAG,CAChB,OAAyB,EACzB,OAA8C,EAC9C,OAAuE;IAgBvE,MAAM,OAAO,GAAG,SAAS,CACvB,OAAO,EACP,CAAC,MAAM,EAAE,EAAE,CAAC,kCAAkC,CAAC,QAAQ,EAAE,CAAC,KAAK,CAAC,MAAM,CAAC,EACvE,yBAAyB,CAC1B,CAAC;IACF,IAAI,CAAC,OAAO,CAAC,EAAE,EAAE,CAAC;QAChB,OAAO,CAAC,OAAO,EAAE,EAAE,MAAM,EAAE,SAAS,EAAE,CAAC,CAAC;IAC1C,CAAC;IACD,MAAM,QAAQ,GAAG,OAAO,CAAC,KAAK,CAAC;IAC/B,MAAM,KAAK,GAAG,IAAI,CAAC;IACnB,MAAM,KAAK,GAAG,UAAU,CAAC,uBAAuB,CAAC,EAAE,CAAC;IACpD,MAAM,MAAM,GAAG,eAAe,CAAC;QAC7B,UAAU,EAAE,QAAQ,EAAE,QAAQ;QAC9B,QAAQ,EAAE,QAAQ,EAAE,MAAM;QAC1B,QAAQ,EAAE,QAAQ,EAAE,MAAM;KAC3B,CAAC,CAAC;IAEH,MAAM,QAAQ,GAAG,IAAI,OAAO,CAAC,UAAU,CAAC;QACtC,MAAM,EAAE,OAAO,EAAE,oBAAoB;eAChC,6CAA6C;KACnD,CAAC,CAAC,CAAC;IACJ,MAAM,aAAa,GAAG,MAAM,eAAe,CAAC,OAAO,CAAC,QAAQ,CAAC,QAAQ,CAAC,CAAC;IACvE,MAAM,eAAe,GAAG,qBAAqB,CAAC,aAAa,CAAC,CAAC;IAE7D,MAAM,OAAO,GAAG;QACd,OAAO,EAAE,OAAO,CAAC,QAAQ;QACzB,OAAO,EAAE,OAAO,EAAE,SAAS,IAAI,OAAO,CAAC,QAAQ,IAAI,EAAE;QACrD,WAAW,EAAE,sBAAsB;QACnC,YAAY,EAAE,IAAI;QAClB,gBAAgB,EAAE,eAAe;QACjC,cAAc,EAAE,OAAO,CAAC,QAAQ,CAAC,QAAQ;QACzC,WAAW,EAAE,OAAO,EAAE,OAAO;eACxB,OAAO,CAAC,QAAQ,CAAC,WAAW;eAC5B,EAAE,QAAQ,EAAE,MAAM,EAAE;QACzB,UAAU,EAAE,OAAO,EAAE,UAAU,IAAI;YACjC,KAAK;YACL,KAAK;YACL,KAAK;YACL,KAAK;YACL,KAAK;SACN;KACF,CAAC;IAEF,MAAM,UAAU,GAAG,OAAO,CAAC,cAAc,CAAC,OAAO,EAAE;QACjD,QAAQ,EAAE,eAAe;QACzB,MAAM,EAAE,KAAK;QACb,OAAO,EAAE,OAAO,EAAE,SAAS;QAC3B,IAAI,EAAE,KAAK;QACX,OAAO,EAAE,QAAQ;QACjB,KAAK,EAAE,MAAM;QACb,IAAI,EAAE,KAAK;QACX,SAAS,EAAE,OAAO,CAAC,QAAQ,CAAC,SAAS;QACrC,SAAS,EAAE,OAAO,EAAE,SAAS,IAAI,OAAO,CAAC,QAAQ,CAAC,SAAS;eACtD,CAAC,CAAC;KACR,EAAE,OAAO,CAAC,CAAC;IACZ,IAAI,CAAC,UAAU,CAAC,EAAE,EAAE,CAAC;QACnB,OAAO,CAAC,UAAU,EAAE,EAAE,MAAM,EAAE,SAAS,EAAE,CAAC,CAAC;IAC7C,CAAC;IACD,MAAM,IAAI,GAAG,UAAU,CAAC,KAAK,CAAC;IAE9B,MAAM,QAAQ,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,IAAI,EAAE;QACvC,OAAO;QACP,UAAU,EAAE,EAAE;QACd,WAAW,EAAE,OAAO,CAAC,WAAW;QAChC,UAAU,EAAE,OAAO,CAAC,UAAU;KAC/B,CAAC,CAAC;IACH,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;QACjB,OAAO,CAAC,QAAQ,EAAE,EAAE,MAAM,EAAE,eAAe,EAAE,OAAO,EAAE,IAAI,EAAE,CAAC,CAAC;IAChE,CAAC;IACD,OAAO,CAAC,QAAQ,EAAE;YAChB,MAAM,EAAE,UAAU;YAClB,SAAS,EAAE,IAAI;YACf,QAAQ,EAAE,QAAQ,CAAC,KAAK;SACzB,CAAC,CAAC;AACL,CAAC"}

package/esm/mcp-server/tools/evalsMonitor.d.ts

@@ -1,7 +0,0 @@
-import { ToolDefinition } from "../tools.js";
-declare const args: {
-    request: import("zod").ZodOptional<import("zod").ZodType<import("../../models/v2evalrunsmonitorop.js").V2EvalRunsMonitorRequest, unknown, import("zod/v4/core").$ZodTypeInternals<import("../../models/v2evalrunsmonitorop.js").V2EvalRunsMonitorRequest, unknown>>>;
-};
-export declare const tool$evalsMonitor: ToolDefinition<typeof args>;
-export {};
-//# sourceMappingURL=evalsMonitor.d.ts.map

package/esm/mcp-server/tools/evalsMonitor.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"evalsMonitor.d.ts","sourceRoot":"","sources":["../../../src/mcp-server/tools/evalsMonitor.ts"],"names":[],"mappings":"AAOA,OAAO,EAAgB,cAAc,EAAE,MAAM,aAAa,CAAC;AAE3D,QAAA,MAAM,IAAI;;CAET,CAAC;AAEF,eAAO,MAAM,iBAAiB,EAAE,cAAc,CAAC,OAAO,IAAI,CA6BzD,CAAC"}

package/esm/mcp-server/tools/evalsMonitor.js

@@ -1,35 +0,0 @@
-/*
- * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
- * @generated-id: f7c3344a66fd
- */
-import { evalsMonitor } from "../../funcs/evalsMonitor.js";
-import { V2EvalRunsMonitorRequest$zodSchema } from "../../models/v2evalrunsmonitorop.js";
-import { formatResult } from "../tools.js";
-const args = {
-    request: V2EvalRunsMonitorRequest$zodSchema.optional(),
-};
-export const tool$evalsMonitor = {
-    name: "evals-monitor",
-    description: `Monitor Runs
-
-Streams an immediate snapshot followed by live status and sanitized activity. Provide \`run_id\` to monitor one public v2 eval run, or omit it to capture all active runs when the connection opens.`,
-    annotations: {
-        "title": "",
-        "destructiveHint": false,
-        "idempotentHint": false,
-        "openWorldHint": false,
-        "readOnlyHint": true,
-    },
-    args,
-    tool: async (client, args, ctx) => {
-        const [result] = await evalsMonitor(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
-        if (!result.ok) {
-            return {
-                content: [{ type: "text", text: result.error.message }],
-                isError: true,
-            };
-        }
-        return formatResult(result.value);
-    },
-};
-//# sourceMappingURL=evalsMonitor.js.map

package/esm/mcp-server/tools/evalsMonitor.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"evalsMonitor.js","sourceRoot":"","sources":["../../../src/mcp-server/tools/evalsMonitor.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,OAAO,EAAE,YAAY,EAAE,MAAM,6BAA6B,CAAC;AAC3D,OAAO,EAAE,kCAAkC,EAAE,MAAM,qCAAqC,CAAC;AACzF,OAAO,EAAE,YAAY,EAAkB,MAAM,aAAa,CAAC;AAE3D,MAAM,IAAI,GAAG;IACX,OAAO,EAAE,kCAAkC,CAAC,QAAQ,EAAE;CACvD,CAAC;AAEF,MAAM,CAAC,MAAM,iBAAiB,GAAgC;IAC5D,IAAI,EAAE,eAAe;IACrB,WAAW,EAAE;;qMAEsL;IACnM,WAAW,EAAE;QACX,OAAO,EAAE,EAAE;QACX,iBAAiB,EAAE,KAAK;QACxB,gBAAgB,EAAE,KAAK;QACvB,eAAe,EAAE,KAAK;QACtB,cAAc,EAAE,IAAI;KACrB;IACD,IAAI;IACJ,IAAI,EAAE,KAAK,EAAE,MAAM,EAAE,IAAI,EAAE,GAAG,EAAE,EAAE;QAChC,MAAM,CAAC,MAAM,CAAC,GAAG,MAAM,YAAY,CACjC,MAAM,EACN,IAAI,CAAC,OAAO,EACZ,EAAE,YAAY,EAAE,EAAE,MAAM,EAAE,GAAG,CAAC,MAAM,EAAE,EAAE,CACzC,CAAC,QAAQ,EAAE,CAAC;QAEb,IAAI,CAAC,MAAM,CAAC,EAAE,EAAE,CAAC;YACf,OAAO;gBACL,OAAO,EAAE,CAAC,EAAE,IAAI,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,CAAC,KAAK,CAAC,OAAO,EAAE,CAAC;gBACvD,OAAO,EAAE,IAAI;aACd,CAAC;QACJ,CAAC;QAED,OAAO,YAAY,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC;IACpC,CAAC;CACF,CAAC"}

package/esm/models/v2evaljudge.d.ts

@@ -1,6 +0,0 @@
-import * as z from "zod";
-export type V2EvalJudge = {
-    criteria: string;
-};
-export declare const V2EvalJudge$zodSchema: z.ZodType<V2EvalJudge>;
-//# sourceMappingURL=v2evaljudge.d.ts.map

package/esm/models/v2evaljudge.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"v2evaljudge.d.ts","sourceRoot":"","sources":["../../src/models/v2evaljudge.ts"],"names":[],"mappings":"AAKA,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AAEzB,MAAM,MAAM,WAAW,GAAG;IAAE,QAAQ,EAAE,MAAM,CAAA;CAAE,CAAC;AAE/C,eAAO,MAAM,qBAAqB,EAAE,CAAC,CAAC,OAAO,CAAC,WAAW,CAEvD,CAAC"}

package/esm/models/v2evaljudge.js

@@ -1,9 +0,0 @@
-/*
- * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
- * @generated-id: ffd5279bc045
- */
-import * as z from "zod";
-export const V2EvalJudge$zodSchema = z.object({
-    criteria: z.string().describe("The outcome that counts as success."),
-});
-//# sourceMappingURL=v2evaljudge.js.map

package/esm/models/v2evaljudge.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"v2evaljudge.js","sourceRoot":"","sources":["../../src/models/v2evaljudge.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AAIzB,MAAM,CAAC,MAAM,qBAAqB,GAA2B,CAAC,CAAC,MAAM,CAAC;IACpE,QAAQ,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CAAC,qCAAqC,CAAC;CACrE,CAAC,CAAC"}

package/esm/models/v2evalrunsmonitorop.d.ts

@@ -1,27 +0,0 @@
-import * as z from "zod";
-import { ClosedEnum } from "../types/enums.js";
-import { HTTPValidationError } from "./httpvalidationerror.js";
-/**
- * Reduced activity or the full sanitized durable trace.
- */
-export declare const Detail: {
-    readonly Summary: "summary";
-    readonly Verbose: "verbose";
-};
-/**
- * Reduced activity or the full sanitized durable trace.
- */
-export type Detail = ClosedEnum<typeof Detail>;
-export declare const Detail$zodSchema: z.ZodEnum<{
-    summary: "summary";
-    verbose: "verbose";
-}>;
-export type V2EvalRunsMonitorRequest = {
-    run_id?: string | undefined;
-    detail?: Detail | undefined;
-    after_id?: number | undefined;
-};
-export declare const V2EvalRunsMonitorRequest$zodSchema: z.ZodType<V2EvalRunsMonitorRequest>;
-export type V2EvalRunsMonitorResponse = string | HTTPValidationError;
-export declare const V2EvalRunsMonitorResponse$zodSchema: z.ZodType<V2EvalRunsMonitorResponse>;
-//# sourceMappingURL=v2evalrunsmonitorop.d.ts.map

package/esm/models/v2evalrunsmonitorop.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"v2evalrunsmonitorop.d.ts","sourceRoot":"","sources":["../../src/models/v2evalrunsmonitorop.ts"],"names":[],"mappings":"AAKA,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AACzB,OAAO,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AAC/C,OAAO,EACL,mBAAmB,EAEpB,MAAM,0BAA0B,CAAC;AAElC;;GAEG;AACH,eAAO,MAAM,MAAM;;;CAGT,CAAC;AACX;;GAEG;AACH,MAAM,MAAM,MAAM,GAAG,UAAU,CAAC,OAAO,MAAM,CAAC,CAAC;AAE/C,eAAO,MAAM,gBAAgB;;;EAGuC,CAAC;AAErE,MAAM,MAAM,wBAAwB,GAAG;IACrC,MAAM,CAAC,EAAE,MAAM,GAAG,SAAS,CAAC;IAC5B,MAAM,CAAC,EAAE,MAAM,GAAG,SAAS,CAAC;IAC5B,QAAQ,CAAC,EAAE,MAAM,GAAG,SAAS,CAAC;CAC/B,CAAC;AAEF,eAAO,MAAM,kCAAkC,EAAE,CAAC,CAAC,OAAO,CACxD,wBAAwB,CAWxB,CAAC;AAEH,MAAM,MAAM,yBAAyB,GAAG,MAAM,GAAG,mBAAmB,CAAC;AAErE,eAAO,MAAM,mCAAmC,EAAE,CAAC,CAAC,OAAO,CACzD,yBAAyB,CAMzB,CAAC"}