@dalmia/calibrate-mcp 0.0.28 → 0.0.30

package/bin/mcp-server.js

@@ -51392,9 +51392,9 @@ var init_config = __esm(() => {
   SDK_METADATA = {
     language: "typescript",
     openapiDocVersion: "0.1.0",
-    sdkVersion: "0.0.28",
+    sdkVersion: "0.0.30",
     genVersion: "2.915.1",
-    userAgent: "speakeasy-sdk/mcp-typescript 0.0.28 2.915.1 0.1.0 @dalmia/calibrate-mcp"
+    userAgent: "speakeasy-sdk/mcp-typescript 0.0.30 2.915.1 0.1.0 @dalmia/calibrate-mcp"
   };
 });
 
@@ -53581,12 +53581,13 @@ var init_routersevaluatorsevaluatorresponse = __esm(() => {
     data_type: RoutersEvaluatorsEvaluatorResponseDataType$zodSchema.describe("The modality the judge reads:\n\n- `text`\n- `audio`\n"),
     description: string2().nullable().optional().describe("What the evaluator checks"),
     evaluator_type: RoutersEvaluatorsEvaluatorResponseEvaluatorType$zodSchema.describe("What the evaluator judges:\n\n- `tts`: TTS audio\n- `stt`: one transcript\n- `llm`: a reply with its conversation history\n- `llm-general`: a standalone input and output pair\n- `conversation`: a full conversation\n"),
-    is_default: boolean2().describe("True for a built-in default evaluator, which you can't edit. False for an evaluator you created, which you can edit and add versions to"),
+    is_default: boolean2().describe("True when the evaluator is a built-in default or your workspace's editable copy of one. False for an evaluator you created yourself"),
     live_version: EvaluatorLiveVersionSummary$zodSchema.nullable().optional().describe("The version that is currently live"),
     live_version_id: string2().nullable().optional().describe("ID of the version that is currently live"),
     name: string2().describe("Evaluator name"),
     output_type: RoutersEvaluatorsEvaluatorResponseOutputType$zodSchema.describe("How the evaluator scores:\n\n- `binary`: pass or fail\n- `rating`: a numeric score, using the scale in `output_config`\n"),
     slug: string2().nullable().optional().describe("Stable slug for a built-in default evaluator"),
+    source_default_slug: string2().nullable().optional().describe("Stable slug of the built-in default this evaluator is your editable copy of. Set on your default forks so you can identify a specific default by it"),
     updated_at: string2().describe("When the evaluator was last updated (ISO 8601 UTC)"),
     uuid: string2().describe("Evaluator ID")
   });
@@ -55866,7 +55867,8 @@ var init_annotationtaskcreate = __esm(() => {
     "stt",
     "llm",
     "llm-general",
-    "conversation"
+    "conversation",
+    "tts"
   ]).describe("Task type. Determines the shape of each item's payload.\n- `stt`: judge a transcript on its own\n- `llm`: judge one response with its conversation history\n- `llm-general`: judge a standalone `input -> output` pair\n- `conversation`: judge a full conversation");
   AnnotationTaskCreate$zodSchema = object({
     description: string2().nullable().optional().describe("A description for the task. Omit for none"),
@@ -56198,7 +56200,8 @@ var init_annotationtaskresponse = __esm(() => {
     "stt",
     "llm",
     "llm-general",
-    "conversation"
+    "conversation",
+    "tts"
   ]).describe("Task type. Determines the shape of each item's payload.\n- `stt`: judge a transcript on its own\n- `llm`: judge one response with its conversation history\n- `llm-general`: judge a standalone `input -> output` pair\n- `conversation`: judge a full conversation");
   AnnotationTaskResponse$zodSchema = object({
     created_at: string2().describe("When the task was created (ISO 8601 UTC)"),
@@ -56750,7 +56753,8 @@ var init_tasksummaryresponse = __esm(() => {
     "stt",
     "llm",
     "llm-general",
-    "conversation"
+    "conversation",
+    "tts"
   ]).describe("Type of the task");
   TaskSummaryResponse$zodSchema = collectExtraKeys(object({
     annotators: array(record(string2(), any())).describe("The annotators who labelled or commented on items in the task, stable across pages"),
@@ -57819,11 +57823,12 @@ var init_evaluatordetailresponsecompact = __esm(() => {
     data_type: EvaluatorDetailResponseCompactDataType$zodSchema.describe("The modality the judge reads:\n\n- `text`\n- `audio`\n"),
     description: string2().nullable().optional().describe("What the evaluator checks"),
     evaluator_type: EvaluatorDetailResponseCompactEvaluatorType$zodSchema.describe("What the evaluator judges:\n\n- `tts`: TTS audio\n- `stt`: one transcript\n- `llm`: a reply with its conversation history\n- `llm-general`: a standalone input and output pair\n- `conversation`: a full conversation\n"),
-    is_default: boolean2().describe("True for a built-in default evaluator, which you can't edit. False for an evaluator you created, which you can edit and add versions to"),
+    is_default: boolean2().describe("True when the evaluator is a built-in default or your workspace's editable copy of one. False for an evaluator you created yourself"),
     live_version_id: string2().nullable().optional().describe("ID of the version that is currently live"),
     name: string2().describe("Evaluator name"),
     output_type: EvaluatorDetailResponseCompactOutputType$zodSchema.describe("How the evaluator scores:\n\n- `binary`: pass or fail\n- `rating`: a numeric score, using the scale in `output_config`\n"),
     slug: string2().nullable().optional().describe("Stable slug for a built-in default evaluator"),
+    source_default_slug: string2().nullable().optional().describe("Stable slug of the built-in default this evaluator is your editable copy of. Set on your default forks so you can identify a specific default by it"),
     updated_at: string2().describe("When the evaluator was last updated (ISO 8601 UTC)"),
     uuid: string2().describe("Evaluator ID"),
     versions: array(EvaluatorVersionCompact$zodSchema).describe("Full version history, oldest first")
@@ -57988,7 +57993,7 @@ var init_listevaluatorsevaluatorsgetop = __esm(() => {
   ListEvaluatorsEvaluatorsGetRequest$zodSchema = object({
     data_type: ListEvaluatorsEvaluatorsGetDataType$zodSchema.nullable().optional().describe("Filter by modality. Omit for all"),
     evaluator_type: ListEvaluatorsEvaluatorsGetEvaluatorType$zodSchema.nullable().optional().describe("Filter by what the evaluator judges. Omit for all types"),
-    include_defaults: boolean2().default(true).describe("When `true`, include the built-in default evaluators alongside the ones you created"),
+    include_defaults: boolean2().default(true).describe("Retained for backward compatibility and no longer filters. Your evaluators, including your editable copies of the defaults, are always returned"),
     limit: int().describe("Maximum number of items to return. Omit for no limit (all items)").nullable().optional(),
     offset: int().default(0).describe("Number of items to skip before returning results"),
     q: string2().describe("Case-insensitive substring search on `name`. Blank is a no-op").nullable().optional(),
@@ -58967,7 +58972,7 @@ fields to change (name, config). Returns the updated test. Fails with
 function createMCPServer(deps) {
   const server = new McpServer({
     name: "CalibrateMcp",
-    version: "0.0.28"
+    version: "0.0.30"
   });
   const getClient = deps.getSDK || (() => new CalibrateMcpCore({
     security: deps.security,
@@ -60194,7 +60199,7 @@ http_headers = { "api-key-auth" = "YOUR_API_KEY_AUTH" }`;
         <h1>Instructions</h1>
         <p>One-click installation for Claude Desktop users</p>
         <div class="instruction-item">
-          <a href="https://github.com/dalmia/calibrate-mcp/releases/download/v0.0.28/mcp-server.mcpb" download="mcp-server.mcpb" class="action-button header-action" style="display: inline-flex; margin-bottom: 16px;">
+          <a href="https://github.com/dalmia/calibrate-mcp/releases/download/v0.0.30/mcp-server.mcpb" download="mcp-server.mcpb" class="action-button header-action" style="display: inline-flex; margin-bottom: 16px;">
             \uD83D\uDCE5 Download MCP Bundle
           </a>
         </div>
@@ -62846,7 +62851,7 @@ var routes = buildRouteMap({
 var app = buildApplication(routes, {
   name: "mcp",
   versionInfo: {
-    currentVersion: "0.0.28"
+    currentVersion: "0.0.30"
   }
 });
 run(app, process4.argv.slice(2), buildContext(process4));
@@ -62854,5 +62859,5 @@ export {
   app
 };
 
-//# debugId=16DFAEE79A0A324C64756E2164756E21
+//# debugId=9864AA851D355E7364756E2164756E21
 //# sourceMappingURL=mcp-server.js.map