@meetkai/mka1 0.54.25 → 0.54.27

package/bin/mcp-server.js

@@ -52917,9 +52917,9 @@ var init_config = __esm(() => {
   SDK_METADATA = {
     language: "typescript",
     openapiDocVersion: "1.1.0",
-    sdkVersion: "0.54.25",
+    sdkVersion: "0.54.27",
     genVersion: "2.935.1",
-    userAgent: "speakeasy-sdk/typescript 0.54.25 2.935.1 1.1.0 @meetkai/mka1"
+    userAgent: "speakeasy-sdk/typescript 0.54.27 2.935.1 1.1.0 @meetkai/mka1"
   };
 });
 
@@ -57121,6 +57121,7 @@ var init_batchendpoint = __esm(() => {
   init_esm();
   BatchEndpoint = {
     RootV1ChatCompletions: "/v1/chat/completions",
+    RootV1Responses: "/v1/responses",
     RootV1Embeddings: "/v1/embeddings",
     RootV1ImagesGenerations: "/v1/images/generations"
   };
@@ -63580,6 +63581,113 @@ var init_evaldataset = __esm(() => {
   });
 });
 
+// src/models/components/evalembeddingconfig.ts
+var Kind, Kind$inboundSchema, Kind$outboundSchema, EvalEmbeddingConfig$inboundSchema, EvalEmbeddingConfig$outboundSchema;
+var init_evalembeddingconfig = __esm(() => {
+  init_esm();
+  init_primitives();
+  init_evaldataset();
+  Kind = {
+    Sts: "sts",
+    Classification: "classification",
+    MultilabelClassification: "multilabel_classification",
+    Retrieval: "retrieval",
+    Reranking: "reranking"
+  };
+  Kind$inboundSchema = nativeEnumType(Kind);
+  Kind$outboundSchema = Kind$inboundSchema;
+  EvalEmbeddingConfig$inboundSchema = objectType({
+    kind: Kind$inboundSchema,
+    text_column: stringType().default("text"),
+    text1_column: stringType().default("sentence1"),
+    text2_column: stringType().default("sentence2"),
+    score_column: stringType().default("score"),
+    label_column: stringType().default("label"),
+    id_column: stringType().default("_id"),
+    title_column: stringType().default("title"),
+    query_id_column: stringType().default("query-id"),
+    corpus_id_column: stringType().default("corpus-id"),
+    corpus_ids_column: stringType().default("corpus-ids"),
+    train_dataset: EvalDataset$inboundSchema.optional(),
+    corpus_dataset: EvalDataset$inboundSchema.optional(),
+    qrels_dataset: EvalDataset$inboundSchema.optional(),
+    top_ranked_dataset: EvalDataset$inboundSchema.optional(),
+    samples_per_label: numberType().int().default(8),
+    n_experiments: numberType().int().default(10),
+    seed: numberType().int().default(42),
+    k_values: arrayType(numberType().int()).optional(),
+    max_chars: numberType().int().default(60000),
+    max_texts: numberType().int().default(50000)
+  }).transform((v2) => {
+    return remap(v2, {
+      text_column: "textColumn",
+      text1_column: "text1Column",
+      text2_column: "text2Column",
+      score_column: "scoreColumn",
+      label_column: "labelColumn",
+      id_column: "idColumn",
+      title_column: "titleColumn",
+      query_id_column: "queryIdColumn",
+      corpus_id_column: "corpusIdColumn",
+      corpus_ids_column: "corpusIdsColumn",
+      train_dataset: "trainDataset",
+      corpus_dataset: "corpusDataset",
+      qrels_dataset: "qrelsDataset",
+      top_ranked_dataset: "topRankedDataset",
+      samples_per_label: "samplesPerLabel",
+      n_experiments: "nExperiments",
+      k_values: "kValues",
+      max_chars: "maxChars",
+      max_texts: "maxTexts"
+    });
+  });
+  EvalEmbeddingConfig$outboundSchema = objectType({
+    kind: Kind$outboundSchema,
+    textColumn: stringType().default("text"),
+    text1Column: stringType().default("sentence1"),
+    text2Column: stringType().default("sentence2"),
+    scoreColumn: stringType().default("score"),
+    labelColumn: stringType().default("label"),
+    idColumn: stringType().default("_id"),
+    titleColumn: stringType().default("title"),
+    queryIdColumn: stringType().default("query-id"),
+    corpusIdColumn: stringType().default("corpus-id"),
+    corpusIdsColumn: stringType().default("corpus-ids"),
+    trainDataset: EvalDataset$outboundSchema.optional(),
+    corpusDataset: EvalDataset$outboundSchema.optional(),
+    qrelsDataset: EvalDataset$outboundSchema.optional(),
+    topRankedDataset: EvalDataset$outboundSchema.optional(),
+    samplesPerLabel: numberType().int().default(8),
+    nExperiments: numberType().int().default(10),
+    seed: numberType().int().default(42),
+    kValues: arrayType(numberType().int()).optional(),
+    maxChars: numberType().int().default(60000),
+    maxTexts: numberType().int().default(50000)
+  }).transform((v2) => {
+    return remap(v2, {
+      textColumn: "text_column",
+      text1Column: "text1_column",
+      text2Column: "text2_column",
+      scoreColumn: "score_column",
+      labelColumn: "label_column",
+      idColumn: "id_column",
+      titleColumn: "title_column",
+      queryIdColumn: "query_id_column",
+      corpusIdColumn: "corpus_id_column",
+      corpusIdsColumn: "corpus_ids_column",
+      trainDataset: "train_dataset",
+      corpusDataset: "corpus_dataset",
+      qrelsDataset: "qrels_dataset",
+      topRankedDataset: "top_ranked_dataset",
+      samplesPerLabel: "samples_per_label",
+      nExperiments: "n_experiments",
+      kValues: "k_values",
+      maxChars: "max_chars",
+      maxTexts: "max_texts"
+    });
+  });
+});
+
 // src/models/components/evalfewshotconfig.ts
 var Strategy, Strategy$inboundSchema, Strategy$outboundSchema, EvalFewshotConfig$inboundSchema, EvalFewshotConfig$outboundSchema;
 var init_evalfewshotconfig = __esm(() => {
@@ -63896,7 +64004,8 @@ var init_evaltasktype = __esm(() => {
     Numeric: "numeric",
     Math: "math",
     Custom: "custom",
-    Transcription: "transcription"
+    Transcription: "transcription",
+    Embedding: "embedding"
   };
   EvalTaskType$inboundSchema = nativeEnumType(EvalTaskType);
   EvalTaskType$outboundSchema = EvalTaskType$inboundSchema;
@@ -63908,6 +64017,7 @@ var init_evaltask = __esm(() => {
   init_esm();
   init_primitives();
   init_evaldataset();
+  init_evalembeddingconfig();
   init_evalfewshotconfig();
   init_evalmetric();
   init_evaloutputextraction();
@@ -63924,7 +64034,8 @@ var init_evaltask = __esm(() => {
     choices: arrayType(stringType()).optional(),
     output_extraction: EvalOutputExtraction$inboundSchema.optional(),
     metrics: arrayType(EvalMetric$inboundSchema).optional(),
-    grader: EvalPythonGrader$inboundSchema,
+    grader: EvalPythonGrader$inboundSchema.optional(),
+    embedding: EvalEmbeddingConfig$inboundSchema.optional(),
     preprocess: EvalPythonPreprocessor$inboundSchema.optional(),
     fewshot: EvalFewshotConfig$inboundSchema.optional(),
     num_fewshot: numberType().int().optional(),
@@ -63948,7 +64059,8 @@ var init_evaltask = __esm(() => {
     choices: arrayType(stringType()).optional(),
     outputExtraction: EvalOutputExtraction$outboundSchema.optional(),
     metrics: arrayType(EvalMetric$outboundSchema).optional(),
-    grader: EvalPythonGrader$outboundSchema,
+    grader: EvalPythonGrader$outboundSchema.optional(),
+    embedding: EvalEmbeddingConfig$outboundSchema.optional(),
     preprocess: EvalPythonPreprocessor$outboundSchema.optional(),
     fewshot: EvalFewshotConfig$outboundSchema.optional(),
     numFewshot: numberType().int().optional(),
@@ -80075,6 +80187,7 @@ var init_components = __esm(() => {
   init_evalartifactsobject();
   init_evaldataset();
   init_evaldatasetformat();
+  init_evalembeddingconfig();
   init_evalfewshotconfig();
   init_evalgenerationconfig();
   init_evalhistoricalimportobject();
@@ -125249,7 +125362,7 @@ var init_llmBatchesCreate2 = __esm(() => {
     name: "llm-batches-create",
     description: `Create a batch
 
-Creates and executes a batch from an uploaded JSONL file of requests. Each line must have custom_id, method, url, and body fields.`,
+Creates and executes a batch from an uploaded JSONL file of requests. Each line must have custom_id, method, url, and body fields. Text batches also require write:responses.`,
     args: args139,
     tool: async (client, args140, ctx) => {
       const [result, apiCall] = await llmBatchesCreate(client, args140.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
@@ -153407,7 +153520,7 @@ Aggregate sandbox usage (session lifecycle, execution, and workspace operations)
 function createMCPServer(deps) {
   const server = new McpServer({
     name: "SDK",
-    version: "0.54.25"
+    version: "0.54.27"
   });
   const client = new SDKCore({
     bearerAuth: deps.bearerAuth,
@@ -155365,7 +155478,7 @@ var routes = rn({
 var app = Ve(routes, {
   name: "mcp",
   versionInfo: {
-    currentVersion: "0.54.25"
+    currentVersion: "0.54.27"
   }
 });
 _t(app, process3.argv.slice(2), buildContext(process3));
@@ -155373,5 +155486,5 @@ export {
   app
 };
 
-//# debugId=ACAA68706DB2011A64756E2164756E21
+//# debugId=1F0AB2945C94B49E64756E2164756E21
 //# sourceMappingURL=mcp-server.js.map