@meetkai/mka1 0.55.33 → 0.55.34

package/bin/mcp-server.js

@@ -72094,9 +72094,9 @@ var init_config = __esm(() => {
   SDK_METADATA = {
     language: "typescript",
     openapiDocVersion: "1.1.0",
-    sdkVersion: "0.55.33",
+    sdkVersion: "0.55.34",
     genVersion: "2.950.2",
-    userAgent: "speakeasy-sdk/typescript 0.55.33 2.950.2 1.1.0 @meetkai/mka1"
+    userAgent: "speakeasy-sdk/typescript 0.55.34 2.950.2 1.1.0 @meetkai/mka1"
   };
 });
 
@@ -90915,6 +90915,47 @@ var init_createdby = __esm(() => {
   });
 });
 
+// src/models/components/evalcandidate.ts
+var EvalCandidate2$inboundSchema, EvalCandidate2$outboundSchema, EvalCandidate1$inboundSchema, EvalCandidate1$outboundSchema, EvalCandidate$inboundSchema, EvalCandidate$outboundSchema;
+var init_evalcandidate = __esm(() => {
+  init_v3();
+  init_primitives();
+  EvalCandidate2$inboundSchema = objectType({
+    type: literalType("saved_agent"),
+    agent_id: stringType(),
+    version: numberType().int()
+  }).transform((v2) => {
+    return remap(v2, {
+      agent_id: "agentId"
+    });
+  });
+  EvalCandidate2$outboundSchema = objectType({
+    type: literalType("saved_agent"),
+    agentId: stringType(),
+    version: numberType().int()
+  }).transform((v2) => {
+    return remap(v2, {
+      agentId: "agent_id"
+    });
+  });
+  EvalCandidate1$inboundSchema = objectType({
+    type: literalType("model"),
+    model: stringType()
+  });
+  EvalCandidate1$outboundSchema = objectType({
+    type: literalType("model"),
+    model: stringType()
+  });
+  EvalCandidate$inboundSchema = unionType([
+    lazyType(() => EvalCandidate1$inboundSchema),
+    lazyType(() => EvalCandidate2$inboundSchema)
+  ]);
+  EvalCandidate$outboundSchema = unionType([
+    lazyType(() => EvalCandidate1$outboundSchema),
+    lazyType(() => EvalCandidate2$outboundSchema)
+  ]);
+});
+
 // src/models/components/evalgenerationrequestconfig.ts
 var Truncation, ServiceTier, EvalGenerationRequestConfigStop$inboundSchema, EvalGenerationRequestConfigStop$outboundSchema, Truncation$inboundSchema, Truncation$outboundSchema, ServiceTier$inboundSchema, ServiceTier$outboundSchema, PrefillThink$inboundSchema, PrefillThink$outboundSchema, EvalGenerationRequestConfig$inboundSchema, EvalGenerationRequestConfig$outboundSchema;
 var init_evalgenerationrequestconfig = __esm(() => {
@@ -91051,11 +91092,13 @@ var CreateEvalRunRequest$inboundSchema, CreateEvalRunRequest$outboundSchema;
 var init_createevalrunrequest = __esm(() => {
   init_v3();
   init_primitives();
+  init_evalcandidate();
   init_evalgenerationrequestconfig();
   CreateEvalRunRequest$inboundSchema = objectType({
     suite_id: stringType(),
     suite_version: numberType().int().optional(),
-    models: arrayType(stringType()),
+    models: arrayType(stringType()).optional(),
+    candidates: arrayType(EvalCandidate$inboundSchema).optional(),
     task_ids: arrayType(stringType()).optional(),
     judge_model: stringType().optional(),
     embedding_model: stringType().optional(),
@@ -91082,7 +91125,8 @@ var init_createevalrunrequest = __esm(() => {
   CreateEvalRunRequest$outboundSchema = objectType({
     suiteId: stringType(),
     suiteVersion: numberType().int().optional(),
-    models: arrayType(stringType()),
+    models: arrayType(stringType()).optional(),
+    candidates: arrayType(EvalCandidate$outboundSchema).optional(),
     taskIds: arrayType(stringType()).optional(),
     judgeModel: stringType().optional(),
     embeddingModel: stringType().optional(),
@@ -91492,20 +91536,24 @@ var init_evalmetric = __esm(() => {
     id: stringType(),
     aggregation: Aggregation$inboundSchema.default("mean"),
     higher_is_better: booleanType().default(true),
+    requires_target: booleanType().optional(),
     description: stringType().optional()
   }).transform((v2) => {
     return remap(v2, {
-      higher_is_better: "higherIsBetter"
+      higher_is_better: "higherIsBetter",
+      requires_target: "requiresTarget"
     });
   });
   EvalMetric$outboundSchema = objectType({
     id: stringType(),
     aggregation: Aggregation$outboundSchema.default("mean"),
     higherIsBetter: booleanType().default(true),
+    requiresTarget: booleanType().optional(),
     description: stringType().optional()
   }).transform((v2) => {
     return remap(v2, {
-      higherIsBetter: "higher_is_better"
+      higherIsBetter: "higher_is_better",
+      requiresTarget: "requires_target"
     });
   });
 });
@@ -95001,6 +95049,48 @@ var init_evalartifactsobject = __esm(() => {
   });
 });
 
+// src/models/components/evalcandidateobject.ts
+var EvalCandidateObject2$inboundSchema, EvalCandidateObject2$outboundSchema, EvalCandidateObject1$inboundSchema, EvalCandidateObject1$outboundSchema, EvalCandidateObject$inboundSchema, EvalCandidateObject$outboundSchema;
+var init_evalcandidateobject = __esm(() => {
+  init_v3();
+  init_primitives();
+  init_discriminatedUnion();
+  EvalCandidateObject2$inboundSchema = objectType({
+    type: literalType("saved_agent"),
+    agent_id: stringType(),
+    version: numberType().int()
+  }).transform((v2) => {
+    return remap(v2, {
+      agent_id: "agentId"
+    });
+  });
+  EvalCandidateObject2$outboundSchema = objectType({
+    type: literalType("saved_agent"),
+    agentId: stringType(),
+    version: numberType().int()
+  }).transform((v2) => {
+    return remap(v2, {
+      agentId: "agent_id"
+    });
+  });
+  EvalCandidateObject1$inboundSchema = objectType({
+    type: literalType("model"),
+    model: stringType()
+  });
+  EvalCandidateObject1$outboundSchema = objectType({
+    type: literalType("model"),
+    model: stringType()
+  });
+  EvalCandidateObject$inboundSchema = discriminatedUnion5("type", {
+    model: lazyType(() => EvalCandidateObject1$inboundSchema),
+    saved_agent: lazyType(() => EvalCandidateObject2$inboundSchema)
+  });
+  EvalCandidateObject$outboundSchema = unionType([
+    lazyType(() => EvalCandidateObject1$outboundSchema),
+    lazyType(() => EvalCandidateObject2$outboundSchema)
+  ]);
+});
+
 // src/models/components/evalhistoricalimportobject.ts
 var EvalHistoricalImportObjectSource$inboundSchema, EvalHistoricalImportObjectSource$outboundSchema, EvalHistoricalImportObject$inboundSchema, EvalHistoricalImportObject$outboundSchema;
 var init_evalhistoricalimportobject = __esm(() => {
@@ -111572,6 +111662,8 @@ var init_components = __esm(() => {
   init_error();
   init_errorevent();
   init_evalartifactsobject();
+  init_evalcandidate();
+  init_evalcandidateobject();
   init_evaldataset();
   init_evaldatasetformat();
   init_evalembeddingconfig();
@@ -123699,6 +123791,7 @@ var init_appendagentevaltrials = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -123751,6 +123844,7 @@ var init_appendagentevaltrials = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -124227,6 +124321,7 @@ var init_cancelevalrun = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -124279,6 +124374,7 @@ var init_cancelevalrun = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -125980,6 +126076,7 @@ var init_createevalrun = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -126032,6 +126129,7 @@ var init_createevalrun = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -131548,6 +131646,7 @@ var init_getevalrun = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -131600,6 +131699,7 @@ var init_getevalrun = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -136445,6 +136545,7 @@ var init_importagentevalrun = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -136497,6 +136598,7 @@ var init_importagentevalrun = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -137874,6 +137976,7 @@ var init_listevalruns = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -137926,6 +138029,7 @@ var init_listevalruns = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -138279,6 +138383,7 @@ var init_listevalscheduleruns = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -138331,6 +138436,7 @@ var init_listevalscheduleruns = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -138912,6 +139018,7 @@ var init_listevalsuiteleaderboardruns = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -138964,6 +139071,7 @@ var init_listevalsuiteleaderboardruns = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -143315,6 +143423,7 @@ var init_rerunfailedevalsamples = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -143367,6 +143476,7 @@ var init_rerunfailedevalsamples = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -143719,6 +143829,7 @@ var init_retryfailedevalrun = __esm(() => {
     team_id: stringType(),
     status: EvalRunStatus$inboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$inboundSchema).optional(),
     task_ids: nullableType(arrayType(stringType())),
     judge_model: nullableType(stringType()),
     embedding_model: nullableType(stringType()),
@@ -143771,6 +143882,7 @@ var init_retryfailedevalrun = __esm(() => {
     teamId: stringType(),
     status: EvalRunStatus$outboundSchema,
     models: arrayType(stringType()),
+    candidates: arrayType(EvalCandidateObject$outboundSchema).optional(),
     taskIds: nullableType(arrayType(stringType())),
     judgeModel: nullableType(stringType()),
     embeddingModel: nullableType(stringType()),
@@ -177738,7 +177850,7 @@ var init_llmEvalsCreateRun2 = __esm(() => {
     name: "llm-evals-create-run",
     description: `Create an eval run
 
-Starts a durable eval run over the selected suite version, tasks, and models.`,
+Starts a durable eval run over the selected suite version and tasks, using model or pinned saved-agent candidates. Supply exactly one of candidates or the legacy models field.`,
     args: args231,
     tool: async (client, args232, ctx) => {
       const [result, apiCall] = await llmEvalsCreateRun(client, args232.request, { fetchOptions: { signal: ctx.mcpReq.signal } }).$inspect();
@@ -208977,7 +209089,7 @@ Aggregate sandbox usage (session lifecycle, execution, and workspace operations)
 function createMCPServer(deps) {
   const server = new McpServer({
     name: "SDK",
-    version: "0.55.33"
+    version: "0.55.34"
   });
   const client = new SDKCore({
     bearerAuth: deps.bearerAuth,
@@ -211178,7 +211290,7 @@ var routes = rn({
 var app = Ve(routes, {
   name: "mcp",
   versionInfo: {
-    currentVersion: "0.55.33"
+    currentVersion: "0.55.34"
   }
 });
 _t(app, process3.argv.slice(2), buildContext(process3));
@@ -211186,5 +211298,5 @@ export {
   app
 };
 
-//# debugId=72FF3E8761A7740C64756E2164756E21
+//# debugId=71FE29807B133ECF64756E2164756E21
 //# sourceMappingURL=mcp-server.js.map