@dalmia/calibrate-mcp 0.0.87 → 0.0.89

package/bin/mcp-server.js

@@ -25997,6 +25997,15 @@ var require_utils2 = __commonJS(function(exports, module) {
     }
     return max;
   };
+  var countElements = function countElements(arr) {
+    var count = 0;
+    for (var i = 0;i < arr.length; ++i) {
+      if (has.call(arr, i)) {
+        count += 1;
+      }
+    }
+    return count;
+  };
   var limitedPair = function limitedPair(first, second, options) {
     var pair = [first, second];
     if (options && typeof options.arrayLimit === "number" && pair.length > options.arrayLimit) {
@@ -26014,7 +26023,7 @@ var require_utils2 = __commonJS(function(exports, module) {
     if (typeof source !== "object" && typeof source !== "function") {
       if (isArray(target)) {
         var nextIndex = target.length;
-        if (options && typeof options.arrayLimit === "number" && nextIndex >= options.arrayLimit) {
+        if (options && typeof options.arrayLimit === "number" && countElements(target) >= options.arrayLimit) {
           if (options.throwOnLimitExceeded) {
             throw new RangeError("Array limit exceeded. Only " + options.arrayLimit + " element" + (options.arrayLimit === 1 ? "" : "s") + " allowed in an array.");
           }
@@ -26083,7 +26092,7 @@ var require_utils2 = __commonJS(function(exports, module) {
           target[i] = item;
         }
       });
-      if (options && typeof options.arrayLimit === "number" && target.length > options.arrayLimit) {
+      if (options && typeof options.arrayLimit === "number" && countElements(target) > options.arrayLimit) {
         if (options.throwOnLimitExceeded) {
           throw new RangeError("Array limit exceeded. Only " + options.arrayLimit + " element" + (options.arrayLimit === 1 ? "" : "s") + " allowed in an array.");
         }
@@ -26577,6 +26586,7 @@ var require_parse = __commonJS(function(exports, module) {
   var has = Object.prototype.hasOwnProperty;
   var isArray = Array.isArray;
   var getPrototypeOf = Object.getPrototypeOf;
+  var fromCharCode = String.fromCharCode;
   var isPlainContainer = function isPlainContainer(value) {
     if (!value || typeof value !== "object") {
       return false;
@@ -26645,7 +26655,12 @@ var require_parse = __commonJS(function(exports, module) {
   };
   var interpretNumericEntities = function(str) {
     return str.replace(/&#(\d+);/g, function($0, numberStr) {
-      return String.fromCharCode(parseInt(numberStr, 10));
+      var c = parseInt(numberStr, 10);
+      if (c > 65535 && c <= 1114111) {
+        c -= 65536;
+        return fromCharCode(55296 | c >> 10, 56320 | c & 1023);
+      }
+      return fromCharCode(c);
     });
   };
   var parseArrayValue = function(val, options, currentArrayLength) {
@@ -26738,6 +26753,9 @@ var require_parse = __commonJS(function(exports, module) {
     }
     return obj;
   };
+  var stripWrappingBrackets = function stripWrappingBrackets(key) {
+    return key.charAt(0) === "[" && key.charAt(key.length - 1) === "]" ? key.slice(1, -1) : key;
+  };
   var parseObject = function(chain, val, options, valuesParsed) {
     var currentArrayLength = 0;
     if (chain.length > 0 && chain[chain.length - 1] === "[]") {
@@ -26756,12 +26774,12 @@ var require_parse = __commonJS(function(exports, module) {
         }
       } else {
         obj = options.plainObjects ? { __proto__: null } : {};
-        var cleanRoot = root.charAt(0) === "[" && root.charAt(root.length - 1) === "]" ? root.slice(1, -1) : root;
+        var cleanRoot = stripWrappingBrackets(root);
         var decodedRoot = options.decodeDotInKeys ? cleanRoot.replace(/%2E/gi, ".") : cleanRoot;
         var index = parseInt(decodedRoot, 10);
         var isValidArrayIndex = !isNaN(index) && root !== decodedRoot && String(index) === decodedRoot && index >= 0 && options.parseArrays;
         if (!options.parseArrays && decodedRoot === "") {
-          obj = { 0: leaf };
+          obj[0] = leaf;
         } else if (isValidArrayIndex && index < options.arrayLimit) {
           obj = [];
           obj[index] = leaf;
@@ -26808,7 +26826,7 @@ var require_parse = __commonJS(function(exports, module) {
   var splitKeyIntoSegments = function splitKeyIntoSegments(originalKey, options) {
     var key = options.allowDots ? topLevelDotsToBrackets(originalKey) : originalKey;
     if (options.depth <= 0) {
-      if (!options.plainObjects && has.call(Object.prototype, key)) {
+      if (!options.plainObjects && has.call(Object.prototype, stripWrappingBrackets(key))) {
         if (!options.allowPrototypes) {
           return;
         }
@@ -27095,10 +27113,11 @@ var require_encodeurl = __commonJS(function(exports, module) {
    */
   module.exports = encodeUrl;
   var ENCODE_CHARS_REGEXP = /(?:[^\x21\x23-\x3B\x3D\x3F-\x5F\x61-\x7A\x7C\x7E]|%(?:[^0-9A-Fa-f]|[0-9A-Fa-f][^0-9A-Fa-f]|$))+/g;
-  var UNMATCHED_SURROGATE_PAIR_REGEXP = /(^|[^\uD800-\uDBFF])[\uDC00-\uDFFF]|[\uD800-\uDBFF]([^\uDC00-\uDFFF]|$)/g;
-  var UNMATCHED_SURROGATE_PAIR_REPLACE = "$1�$2";
+  var SURROGATE_REGEXP = /([\uD800-\uDBFF][\uDC00-\uDFFF])|[\uD800-\uDFFF]/g;
   function encodeUrl(url) {
-    return String(url).replace(UNMATCHED_SURROGATE_PAIR_REGEXP, UNMATCHED_SURROGATE_PAIR_REPLACE).replace(ENCODE_CHARS_REGEXP, encodeURI);
+    return String(url).replace(SURROGATE_REGEXP, function(_, pair) {
+      return pair || "�";
+    }).replace(ENCODE_CHARS_REGEXP, encodeURI);
   }
 });
 
@@ -53279,9 +53298,9 @@ var init_config = __esm(() => {
   SDK_METADATA = {
     language: "typescript",
     openapiDocVersion: "0.1.0",
-    sdkVersion: "0.0.87",
+    sdkVersion: "0.0.89",
     genVersion: "2.915.1",
-    userAgent: "speakeasy-sdk/mcp-typescript 0.0.87 2.915.1 0.1.0 @dalmia/calibrate-mcp"
+    userAgent: "speakeasy-sdk/mcp-typescript 0.0.89 2.915.1 0.1.0 @dalmia/calibrate-mcp"
   };
 });
 
@@ -61851,6 +61870,146 @@ var init_evaluatorsList2 = __esm(() => {
   };
 });
 
+// src/models/currentworkspaceresponse.ts
+var CurrentWorkspaceResponse$zodSchema;
+var init_currentworkspaceresponse = __esm(() => {
+  init_zod();
+  CurrentWorkspaceResponse$zodSchema = object({
+    name: string2().describe("Workspace display name"),
+    uuid: string2().describe("Workspace ID. Put this at the front of a Calibrate web address to link straight to something in this workspace, as in `/<uuid>/tests`")
+  });
+});
+
+// src/models/getcurrentworkspaceorganizationscurrentgetop.ts
+var GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema, GetCurrentWorkspaceOrganizationsCurrentGetResponse$zodSchema;
+var init_getcurrentworkspaceorganizationscurrentgetop = __esm(() => {
+  init_zod();
+  init_currentworkspaceresponse();
+  init_httpvalidationerror();
+  GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema = object({
+    xAPIKey: string2().nullable().optional()
+  });
+  GetCurrentWorkspaceOrganizationsCurrentGetResponse$zodSchema = union([
+    CurrentWorkspaceResponse$zodSchema,
+    HTTPValidationError$zodSchema
+  ]);
+});
+
+// src/funcs/organizationsCurrent.ts
+function organizationsCurrent(client$, request, options) {
+  return new APIPromise($do40(client$, request, options));
+}
+async function $do40(client$, request, options) {
+  const parsed$ = safeParse4(request, (value$) => GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema.optional().parse(value$), "Input validation failed");
+  if (!parsed$.ok) {
+    return [parsed$, { status: "invalid" }];
+  }
+  const payload$ = parsed$.value;
+  const body$ = null;
+  const path$ = pathToFunc("/organizations/current")();
+  const headers$ = new Headers(compactMap({
+    Accept: "application/json",
+    "X-API-Key": encodeSimple("X-API-Key", payload$?.xAPIKey, {
+      explode: false,
+      charEncoding: "none"
+    })
+  }));
+  const securityInput = await extractSecurity(client$._options.security);
+  const requestSecurity = resolveGlobalSecurity(securityInput);
+  const context = {
+    options: client$._options,
+    baseURL: options?.serverURL ?? client$._baseURL ?? "",
+    operationID: "get_current_workspace_organizations_current_get",
+    oAuth2Scopes: null,
+    resolvedSecurity: requestSecurity,
+    securitySource: client$._options.security,
+    retryConfig: options?.retries || client$._options.retryConfig || { strategy: "none" },
+    retryCodes: options?.retryCodes || [
+      "429",
+      "500",
+      "502",
+      "503",
+      "504"
+    ]
+  };
+  const requestRes = client$._createRequest(context, {
+    security: requestSecurity,
+    method: "GET",
+    baseURL: options?.serverURL,
+    path: path$,
+    headers: headers$,
+    body: body$,
+    userAgent: client$._options.userAgent,
+    timeoutMs: options?.timeoutMs || client$._options.timeoutMs || -1
+  }, options);
+  if (!requestRes.ok) {
+    return [requestRes, { status: "invalid" }];
+  }
+  const req$ = requestRes.value;
+  const doResult = await client$._do(req$, {
+    context,
+    errorCodes: [],
+    retryConfig: context.retryConfig,
+    retryCodes: context.retryCodes
+  });
+  if (!doResult.ok) {
+    return [doResult, { status: "request-error", request: req$ }];
+  }
+  return [doResult, {
+    status: "complete",
+    request: req$,
+    response: doResult.value
+  }];
+}
+var init_organizationsCurrent = __esm(() => {
+  init_encodings();
+  init_primitives();
+  init_schemas4();
+  init_security();
+  init_url();
+  init_getcurrentworkspaceorganizationscurrentgetop();
+  init_async();
+});
+
+// src/mcp-server/tools/organizationsCurrent.ts
+var args40, tool$organizationsCurrent;
+var init_organizationsCurrent2 = __esm(() => {
+  init_organizationsCurrent();
+  init_getcurrentworkspaceorganizationscurrentgetop();
+  init_tools();
+  args40 = {
+    request: GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema.optional()
+  };
+  tool$organizationsCurrent = {
+    name: "get-current-workspace",
+    description: `Get the workspace the caller's API key belongs to, as its ID and
+name. Put the ID at the front of a Calibrate web address to link
+straight to something in that workspace, as in \`/<uuid>/tests\`.
+Without it, a link opens whichever workspace the reader looked at
+last. Requires a Calibrate API key (CALIBRATE_API_KEY).
+`,
+    scopes: ["read"],
+    annotations: {
+      title: "Get the current workspace",
+      destructiveHint: false,
+      idempotentHint: true,
+      openWorldHint: true,
+      readOnlyHint: true
+    },
+    args: args40,
+    tool: async (client, args, ctx) => {
+      const [result] = await organizationsCurrent(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
+      if (!result.ok) {
+        return {
+          content: [{ type: "text", text: result.error.message }],
+          isError: true
+        };
+      }
+      return formatResult(result.value);
+    }
+  };
+});
+
 // src/models/chatmessage.ts
 var Role$zodSchema, ChatMessage$zodSchema;
 var init_chatmessage = __esm(() => {
@@ -61957,9 +62116,9 @@ var init_bulkuploadteststestsbulkpostop = __esm(() => {
 
 // src/funcs/testsBulkCreate.ts
 function testsBulkCreate(client$, request, options) {
-  return new APIPromise($do40(client$, request, options));
+  return new APIPromise($do41(client$, request, options));
 }
-async function $do40(client$, request, options) {
+async function $do41(client$, request, options) {
   const parsed$ = safeParse4(request, (value$) => BulkUploadTestsTestsBulkPostRequest$zodSchema.parse(value$), "Input validation failed");
   if (!parsed$.ok) {
     return [parsed$, { status: "invalid" }];
@@ -62033,12 +62192,12 @@ var init_testsBulkCreate = __esm(() => {
 });
 
 // src/mcp-server/tools/testsBulkCreate.ts
-var args40, tool$testsBulkCreate;
+var args41, tool$testsBulkCreate;
 var init_testsBulkCreate2 = __esm(() => {
   init_testsBulkCreate();
   init_bulkuploadteststestsbulkpostop();
   init_tools();
-  args40 = {
+  args41 = {
     request: BulkUploadTestsTestsBulkPostRequest$zodSchema
   };
   tool$testsBulkCreate = {
@@ -62055,7 +62214,7 @@ tests including their UUIDs.
       openWorldHint: true,
       readOnlyHint: false
     },
-    args: args40,
+    args: args41,
     tool: async (client, args, ctx) => {
       const [result] = await testsBulkCreate(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
       if (!result.ok) {
@@ -62117,9 +62276,9 @@ var init_createtestendpointtestspostop = __esm(() => {
 
 // src/funcs/testsCreate.ts
 function testsCreate(client$, request, options) {
-  return new APIPromise($do41(client$, request, options));
+  return new APIPromise($do42(client$, request, options));
 }
-async function $do41(client$, request, options) {
+async function $do42(client$, request, options) {
   const parsed$ = safeParse4(request, (value$) => CreateTestEndpointTestsPostRequest$zodSchema.parse(value$), "Input validation failed");
   if (!parsed$.ok) {
     return [parsed$, { status: "invalid" }];
@@ -62193,12 +62352,12 @@ var init_testsCreate = __esm(() => {
 });
 
 // src/mcp-server/tools/testsCreate.ts
-var args41, tool$testsCreate;
+var args42, tool$testsCreate;
 var init_testsCreate2 = __esm(() => {
   init_testsCreate();
   init_createtestendpointtestspostop();
   init_tools();
-  args41 = {
+  args42 = {
     request: CreateTestEndpointTestsPostRequest$zodSchema
   };
   tool$testsCreate = {
@@ -62215,7 +62374,7 @@ Requires a Calibrate API key (CALIBRATE_API_KEY).
       openWorldHint: true,
       readOnlyHint: false
     },
-    args: args41,
+    args: args42,
     tool: async (client, args, ctx) => {
       const [result] = await testsCreate(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
       if (!result.ok) {
@@ -62268,9 +62427,9 @@ var init_gettestendpointteststestuuidgetop = __esm(() => {
 
 // src/funcs/testsGet.ts
 function testsGet(client$, request, options) {
-  return new APIPromise($do42(client$, request, options));
+  return new APIPromise($do43(client$, request, options));
 }
-async function $do42(client$, request, options) {
+async function $do43(client$, request, options) {
   const parsed$ = safeParse4(request, (value$) => GetTestEndpointTestsTestUuidGetRequest$zodSchema.parse(value$), "Input validation failed");
   if (!parsed$.ok) {
     return [parsed$, { status: "invalid" }];
@@ -62349,12 +62508,12 @@ var init_testsGet = __esm(() => {
 });
 
 // src/mcp-server/tools/testsGet.ts
-var args42, tool$testsGet;
+var args43, tool$testsGet;
 var init_testsGet2 = __esm(() => {
   init_testsGet();
   init_gettestendpointteststestuuidgetop();
   init_tools();
-  args42 = {
+  args43 = {
     request: GetTestEndpointTestsTestUuidGetRequest$zodSchema
   };
   tool$testsGet = {
@@ -62371,7 +62530,7 @@ exist or belongs to another org.
       openWorldHint: true,
       readOnlyHint: true
     },
-    args: args42,
+    args: args43,
     tool: async (client, args, ctx) => {
       const [result] = await testsGet(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
       if (!result.ok) {
@@ -62405,9 +62564,9 @@ var init_listteststestsgetop = __esm(() => {
 
 // src/funcs/testsList.ts
 function testsList(client$, request, options) {
-  return new APIPromise($do43(client$, request, options));
+  return new APIPromise($do44(client$, request, options));
 }
-async function $do43(client$, request, options) {
+async function $do44(client$, request, options) {
   const parsed$ = safeParse4(request, (value$) => ListTestsTestsGetRequest$zodSchema.optional().parse(value$), "Input validation failed");
   if (!parsed$.ok) {
     return [parsed$, { status: "invalid" }];
@@ -62486,12 +62645,12 @@ var init_testsList = __esm(() => {
 });
 
 // src/mcp-server/tools/testsList.ts
-var args43, tool$testsList;
+var args44, tool$testsList;
 var init_testsList2 = __esm(() => {
   init_testsList();
   init_listteststestsgetop();
   init_tools();
-  args43 = {
+  args44 = {
     request: ListTestsTestsGetRequest$zodSchema.optional()
   };
   tool$testsList = {
@@ -62508,7 +62667,7 @@ before creating runs.
       openWorldHint: true,
       readOnlyHint: true
     },
-    args: args43,
+    args: args44,
     tool: async (client, args, ctx) => {
       const [result] = await testsList(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
       if (!result.ok) {
@@ -62561,9 +62720,9 @@ var init_updatetestendpointteststestuuidputop = __esm(() => {
 
 // src/funcs/testsUpdate.ts
 function testsUpdate(client$, request, options) {
-  return new APIPromise($do44(client$, request, options));
+  return new APIPromise($do45(client$, request, options));
 }
-async function $do44(client$, request, options) {
+async function $do45(client$, request, options) {
   const parsed$ = safeParse4(request, (value$) => UpdateTestEndpointTestsTestUuidPutRequest$zodSchema.parse(value$), "Input validation failed");
   if (!parsed$.ok) {
     return [parsed$, { status: "invalid" }];
@@ -62643,12 +62802,12 @@ var init_testsUpdate = __esm(() => {
 });
 
 // src/mcp-server/tools/testsUpdate.ts
-var args44, tool$testsUpdate;
+var args45, tool$testsUpdate;
 var init_testsUpdate2 = __esm(() => {
   init_testsUpdate();
   init_updatetestendpointteststestuuidputop();
   init_tools();
-  args44 = {
+  args45 = {
     request: UpdateTestEndpointTestsTestUuidPutRequest$zodSchema
   };
   tool$testsUpdate = {
@@ -62665,7 +62824,7 @@ fields to change (name, config). Returns the updated test. Fails with
       openWorldHint: true,
       readOnlyHint: false
     },
-    args: args44,
+    args: args45,
     tool: async (client, args, ctx) => {
       const [result] = await testsUpdate(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
       if (!result.ok) {
@@ -62789,9 +62948,9 @@ var init_ingesttracetracespostop = __esm(() => {
 
 // src/funcs/tracesCreate.ts
 function tracesCreate(client$, request, options) {
-  return new APIPromise($do45(client$, request, options));
+  return new APIPromise($do46(client$, request, options));
 }
-async function $do45(client$, request, options) {
+async function $do46(client$, request, options) {
   const parsed$ = safeParse4(request, (value$) => IngestTraceTracesPostRequest$zodSchema.parse(value$), "Input validation failed");
   if (!parsed$.ok) {
     return [parsed$, { status: "invalid" }];
@@ -62865,12 +63024,12 @@ var init_tracesCreate = __esm(() => {
 });
 
 // src/mcp-server/tools/tracesCreate.ts
-var args45, tool$tracesCreate;
+var args46, tool$tracesCreate;
 var init_tracesCreate2 = __esm(() => {
   init_tracesCreate();
   init_ingesttracetracespostop();
   init_tools();
-  args45 = {
+  args46 = {
     request: IngestTraceTracesPostRequest$zodSchema
   };
   tool$tracesCreate = {
@@ -62889,7 +63048,7 @@ hits its trace limit.
       openWorldHint: true,
       readOnlyHint: false
     },
-    args: args45,
+    args: args46,
     tool: async (client, args, ctx) => {
       const [result] = await tracesCreate(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
       if (!result.ok) {
@@ -62903,43 +63062,188 @@ hits its trace limit.
   };
 });
 
-// src/models/currentworkspaceresponse.ts
-var CurrentWorkspaceResponse$zodSchema;
-var init_currentworkspaceresponse = __esm(() => {
+// src/models/tracescoreaverage.ts
+var TraceScoreAverageOutputType$zodSchema, TraceScoreAverage$zodSchema;
+var init_tracescoreaverage = __esm(() => {
   init_zod();
-  CurrentWorkspaceResponse$zodSchema = object({
-    name: string2().describe("Workspace display name"),
-    uuid: string2().describe("Workspace ID. Put this at the front of a Calibrate web address to link straight to something in this workspace, as in `/<uuid>/tests`")
+  TraceScoreAverageOutputType$zodSchema = _enum([
+    "binary",
+    "rating"
+  ]).describe("How the evaluator scores:\n\n- `binary`: pass or fail\n- `rating`: a numeric score, using the scale in `output_config`\n");
+  TraceScoreAverage$zodSchema = object({
+    average: number2().describe("Mean of those scores. A binary evaluator averages 0 and 1, so its mean is the share that passed"),
+    evaluator_uuid: string2().describe("ID of the evaluator these numbers are for"),
+    name: string2().describe("Name of the evaluator. Uses the last stored name if it has been deleted"),
+    output_type: TraceScoreAverageOutputType$zodSchema.describe("How the evaluator scores:\n\n- `binary`: pass or fail\n- `rating`: a numeric score, using the scale in `output_config`\n"),
+    scale_max: number2().nullable().optional().describe("Highest value on a rating scale"),
+    scale_min: number2().nullable().optional().describe("Lowest value on a rating scale"),
+    traces_scored: int().describe("How many of the matching traces this evaluator scored")
   });
 });
 
-// src/models/getcurrentworkspaceorganizationscurrentgetop.ts
-var GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema, GetCurrentWorkspaceOrganizationsCurrentGetResponse$zodSchema;
-var init_getcurrentworkspaceorganizationscurrentgetop = __esm(() => {
+// src/models/traceevalrunstatus.ts
+var TraceEvalRunStatus$zodSchema;
+var init_traceevalrunstatus = __esm(() => {
+  init_zod();
+  TraceEvalRunStatus$zodSchema = _enum([
+    "pending",
+    "processing",
+    "completed",
+    "failed",
+    "skipped"
+  ]);
+});
+
+// src/models/tracescoreresult.ts
+var TraceScoreResultEvaluatorType$zodSchema, TraceScoreResultOutputType$zodSchema, TraceScoreResult$zodSchema;
+var init_tracescoreresult = __esm(() => {
+  init_zod();
+  TraceScoreResultEvaluatorType$zodSchema = _enum([
+    "tts",
+    "stt",
+    "llm",
+    "llm-general",
+    "conversation",
+    "tool-call"
+  ]);
+  TraceScoreResultOutputType$zodSchema = _enum([
+    "binary",
+    "rating"
+  ]).describe("How the evaluator scores:\n\n- `binary`: pass or fail\n- `rating`: a numeric score, using the scale in `output_config`\n");
+  TraceScoreResult$zodSchema = object({
+    evaluator_type: TraceScoreResultEvaluatorType$zodSchema.nullable().optional().describe("What the evaluator judges:\n\n- `tts`: TTS audio\n- `stt`: one transcript\n- `llm`: a reply with its conversation history\n- `llm-general`: a standalone input and output pair\n- `conversation`: a full conversation\n- `tool-call`: whether the agent called the right tool, labelled by a person\n"),
+    evaluator_uuid: string2().describe("ID of the evaluator that produced this result"),
+    evaluator_version_id: string2().describe("ID of the evaluator version that produced this result"),
+    name: string2().describe("Name of the evaluator. Uses the last stored name if it has been deleted"),
+    output_type: TraceScoreResultOutputType$zodSchema.describe("How the evaluator scores:\n\n- `binary`: pass or fail\n- `rating`: a numeric score, using the scale in `output_config`\n"),
+    passed: boolean2().describe("Whether this evaluator passed. A binary result passes on 1. A rating passes at the top of its scale"),
+    reasoning: string2().nullable().optional().describe("The judge's rationale for this result"),
+    scale_max: number2().nullable().optional().describe("Highest value on a rating scale"),
+    scale_min: number2().nullable().optional().describe("Lowest value on a rating scale"),
+    value: number2().describe("The judged result. Binary stores 0 or 1. Rating stores the numeric score")
+  });
+});
+
+// src/models/tracesummary.ts
+var TraceSummaryInput$zodSchema, TraceSummary$zodSchema;
+var init_tracesummary = __esm(() => {
+  init_zod();
+  init_traceevalrunstatus();
+  init_tracemetadataentry();
+  init_traceoutput();
+  init_tracescoreresult();
+  init_tracetoolcall();
+  init_traceturn();
+  TraceSummaryInput$zodSchema = union([
+    string2(),
+    array(TraceTurn$zodSchema)
+  ]).describe("What the agent was given for this turn. For a `general` agent, the standalone prompt as a string. For a `conversation` agent, the history up to the reported output, oldest turn first, in OpenAI chat format. Sent only with `include_content`");
+  TraceSummary$zodSchema = object({
+    agent_id: string2().describe("ID of the agent that produced the turn"),
+    conversation_id: string2().nullable().optional().describe("The conversation ID you sent, if any"),
+    created_at: string2().describe("When the trace was created (ISO 8601 UTC)"),
+    input: union([
+      string2(),
+      array(TraceTurn$zodSchema)
+    ]).nullable().optional().describe("What the agent was given for this turn. For a `general` agent, the standalone prompt as a string. For a `conversation` agent, the history up to the reported output, oldest turn first, in OpenAI chat format. Sent only with `include_content`"),
+    input_preview: string2().nullable().optional().describe("The standalone prompt or the last user message, truncated for display"),
+    labels: array(string2()).describe("Your own tags for this turn, such as an environment or a release. Matched exactly when filtering, so keep the spelling stable"),
+    latest_run_error: string2().nullable().optional().describe("Why the latest scoring run was skipped or failed"),
+    latest_run_status: TraceEvalRunStatus$zodSchema.nullable().optional().describe("Status of the latest scoring run for this trace"),
+    message_id: string2().nullable().optional().describe("The message ID you sent, if any"),
+    metadata: array(TraceMetadataEntry$zodSchema).nullable().optional().describe("Key-value pairs stored with the trace. Sent only with `include_content`"),
+    metadata_count: int().describe("Number of metadata entries stored with the trace"),
+    output: TraceOutput$zodSchema.nullable().optional().describe("What the agent produced for this turn. Sent only with `include_content`"),
+    response_preview: string2().nullable().optional().describe("The agent reply, truncated for display"),
+    results: array(TraceScoreResult$zodSchema).optional().describe("Results for each evaluator on the latest scoring run"),
+    tool_call_count: int().describe("Number of tool calls the agent issued for this turn"),
+    tool_calls: array(TraceToolCall$zodSchema).describe("Tools the agent issued on this turn, with the arguments it passed"),
+    tool_names: array(string2()).describe("Names of the tools the agent issued on this turn, in order"),
+    turn_count: int().describe("Number of turns in the stored input. A standalone prompt counts as 1"),
+    uuid: string2().describe("Unique ID for the trace")
+  });
+});
+
+// src/models/tracelistresponse.ts
+var TraceListResponse$zodSchema;
+var init_tracelistresponse = __esm(() => {
+  init_zod();
+  init_tracescoreaverage();
+  init_tracesummary();
+  TraceListResponse$zodSchema = object({
+    items: array(TraceSummary$zodSchema).describe("The page of results"),
+    limit: int().nullable().optional().describe("The applied `limit` (null when unbounded)"),
+    offset: int().default(0).describe("The applied `offset`"),
+    score_averages: array(TraceScoreAverage$zodSchema).nullable().optional().describe("One entry per evaluator that scored at least one matching trace, averaged over every matching trace rather than this page. Ask for it with `include_score_averages`"),
+    total: int().describe("Total number of items matching the query, before pagination")
+  }).describe("The page, plus the averages when `include_score_averages` asked for them.");
+});
+
+// src/models/listtracesendpointtracesgetop.ts
+var ListTracesEndpointTracesGetOutputType$zodSchema, SortOrder$zodSchema, ListTracesEndpointTracesGetRequest$zodSchema, ListTracesEndpointTracesGetResponse$zodSchema;
+var init_listtracesendpointtracesgetop = __esm(() => {
   init_zod();
-  init_currentworkspaceresponse();
   init_httpvalidationerror();
-  GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema = object({
+  init_tracelistresponse();
+  ListTracesEndpointTracesGetOutputType$zodSchema = _enum([
+    "response",
+    "tool_call"
+  ]).describe("Return only traces whose output is of this kind. `response` covers every trace carrying a reply, including one that also issued tool calls. `tool_call` covers traces that only issued tool calls");
+  SortOrder$zodSchema = _enum([
+    "asc",
+    "desc"
+  ]).describe("Whether `sort_by_evaluator` puts the highest scores first or the lowest. Ignored without it");
+  ListTracesEndpointTracesGetRequest$zodSchema = object({
+    agent_id: string2().describe("Return only traces from this agent").nullable().optional(),
+    conversation_id: string2().describe("Return only the turns of this conversation. Requires `agent_id`").nullable().optional(),
+    created_after: datetime2({ offset: true }).describe("Return only traces created at or after this time (ISO 8601). A time with no timezone is read as UTC").nullable().optional(),
+    created_before: datetime2({ offset: true }).describe("Return only traces created before this time (ISO 8601). A time with no timezone is read as UTC").nullable().optional(),
+    include_content: boolean2().default(false).describe("Also return each trace's full input, output and metadata. Use `limit` and `offset` to page through them"),
+    include_score_averages: boolean2().default(false).describe("Also return each evaluator's mean score over every matching trace, not just this page. Costs a pass over those traces, so ask for it when the filters change rather than on every page"),
+    labels: array(string2()).describe("Return only traces carrying at least one of these labels. Repeat the parameter for each label").nullable().optional(),
+    limit: int().default(50).describe("Maximum number of items to return"),
+    offset: int().default(0).describe("Number of items to skip before returning results"),
+    output_type: ListTracesEndpointTracesGetOutputType$zodSchema.nullable().optional().describe("Return only traces whose output is of this kind. `response` covers every trace carrying a reply, including one that also issued tool calls. `tool_call` covers traces that only issued tool calls"),
+    q: string2().describe("Return only traces containing this text in their message ID, conversation ID, conversation history, output, or metadata").nullable().optional(),
+    score: array(string2()).describe("Return only traces whose latest scoring run matches this condition. Each is written as `<evaluator ID>:<condition>`, where the condition is `passed`, `failed`, or one of `=`, `!=`, `<`, `<=`, `>`, `>=` followed by a number, such as `>=4`. Repeat the parameter to require every condition at once").nullable().optional(),
+    sort_by_evaluator: string2().describe("Order traces by this evaluator's score on each trace's latest scoring run. Traces it has not scored come last whichever direction you pick").nullable().optional(),
+    sort_order: SortOrder$zodSchema.default("desc").describe("Whether `sort_by_evaluator` puts the highest scores first or the lowest. Ignored without it"),
     xAPIKey: string2().nullable().optional()
   });
-  GetCurrentWorkspaceOrganizationsCurrentGetResponse$zodSchema = union([
-    CurrentWorkspaceResponse$zodSchema,
+  ListTracesEndpointTracesGetResponse$zodSchema = union([
+    TraceListResponse$zodSchema,
     HTTPValidationError$zodSchema
   ]);
 });
 
-// src/funcs/workspaceCurrent.ts
-function workspaceCurrent(client$, request, options) {
-  return new APIPromise($do46(client$, request, options));
+// src/funcs/tracesList.ts
+function tracesList(client$, request, options) {
+  return new APIPromise($do47(client$, request, options));
 }
-async function $do46(client$, request, options) {
-  const parsed$ = safeParse4(request, (value$) => GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema.optional().parse(value$), "Input validation failed");
+async function $do47(client$, request, options) {
+  const parsed$ = safeParse4(request, (value$) => ListTracesEndpointTracesGetRequest$zodSchema.optional().parse(value$), "Input validation failed");
   if (!parsed$.ok) {
     return [parsed$, { status: "invalid" }];
   }
   const payload$ = parsed$.value;
   const body$ = null;
-  const path$ = pathToFunc("/organizations/current")();
+  const path$ = pathToFunc("/traces")();
+  const query$ = encodeFormQuery({
+    agent_id: payload$?.agent_id,
+    conversation_id: payload$?.conversation_id,
+    created_after: payload$?.created_after,
+    created_before: payload$?.created_before,
+    include_content: payload$?.include_content,
+    include_score_averages: payload$?.include_score_averages,
+    labels: payload$?.labels,
+    limit: payload$?.limit,
+    offset: payload$?.offset,
+    output_type: payload$?.output_type,
+    q: payload$?.q,
+    score: payload$?.score,
+    sort_by_evaluator: payload$?.sort_by_evaluator,
+    sort_order: payload$?.sort_order
+  });
   const headers$ = new Headers(compactMap({
     Accept: "application/json",
     "X-API-Key": encodeSimple("X-API-Key", payload$?.xAPIKey, {
@@ -62952,7 +63256,7 @@ async function $do46(client$, request, options) {
   const context = {
     options: client$._options,
     baseURL: options?.serverURL ?? client$._baseURL ?? "",
-    operationID: "get_current_workspace_organizations_current_get",
+    operationID: "list_traces_endpoint_traces_get",
     oAuth2Scopes: null,
     resolvedSecurity: requestSecurity,
     securitySource: client$._options.security,
@@ -62971,6 +63275,7 @@ async function $do46(client$, request, options) {
     baseURL: options?.serverURL,
     path: path$,
     headers: headers$,
+    query: query$,
     body: body$,
     userAgent: client$._options.userAgent,
     timeoutMs: options?.timeoutMs || client$._options.timeoutMs || -1
@@ -62994,44 +63299,46 @@ async function $do46(client$, request, options) {
     response: doResult.value
   }];
 }
-var init_workspaceCurrent = __esm(() => {
+var init_tracesList = __esm(() => {
   init_encodings();
   init_primitives();
   init_schemas4();
   init_security();
   init_url();
-  init_getcurrentworkspaceorganizationscurrentgetop();
+  init_listtracesendpointtracesgetop();
   init_async();
 });
 
-// src/mcp-server/tools/workspaceCurrent.ts
-var args46, tool$workspaceCurrent;
-var init_workspaceCurrent2 = __esm(() => {
-  init_workspaceCurrent();
-  init_getcurrentworkspaceorganizationscurrentgetop();
+// src/mcp-server/tools/tracesList.ts
+var args47, tool$tracesList;
+var init_tracesList2 = __esm(() => {
+  init_tracesList();
+  init_listtracesendpointtracesgetop();
   init_tools();
-  args46 = {
-    request: GetCurrentWorkspaceOrganizationsCurrentGetRequest$zodSchema.optional()
-  };
-  tool$workspaceCurrent = {
-    name: "get-current-workspace",
-    description: `Get the workspace the caller's API key belongs to, as its ID and
-name. Put the ID at the front of a Calibrate web address to link
-straight to something in that workspace, as in \`/<uuid>/tests\`.
-Without it, a link opens whichever workspace the reader looked at
-last. Requires a Calibrate API key (CALIBRATE_API_KEY).
+  args47 = {
+    request: ListTracesEndpointTracesGetRequest$zodSchema.optional()
+  };
+  tool$tracesList = {
+    name: "list-traces",
+    description: `List the traces in the caller's org, newest first, one page at a
+time. Each trace carries a short preview of its last user message
+and reply, its tool calls and labels, and the scores from its
+latest scoring run with each evaluator's reasoning and evaluator
+version. Filter by agent, label, creation time, text, or score.
+Set include_content to also get each trace's full input, output
+and metadata.
 `,
     scopes: ["read"],
     annotations: {
-      title: "Get the current workspace",
+      title: "List traces",
       destructiveHint: false,
-      idempotentHint: true,
+      idempotentHint: false,
       openWorldHint: true,
       readOnlyHint: true
     },
-    args: args46,
+    args: args47,
     tool: async (client, args, ctx) => {
-      const [result] = await workspaceCurrent(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
+      const [result] = await tracesList(client, args.request, { fetchOptions: { signal: ctx.signal } }).$inspect();
       if (!result.ok) {
         return {
           content: [{ type: "text", text: result.error.message }],
@@ -63047,7 +63354,7 @@ last. Requires a Calibrate API key (CALIBRATE_API_KEY).
 function createMCPServer(deps) {
   const server = new McpServer({
     name: "CalibrateMcp",
-    version: "0.0.87"
+    version: "0.0.89"
   });
   const getClient = deps.getSDK || (() => new CalibrateMcpCore({
     security: deps.security,
@@ -63110,8 +63417,9 @@ function createMCPServer(deps) {
   tool(tool$annotatorsCreate);
   tool(tool$annotatorsList);
   tool(tool$annotatorsUpdate);
-  tool(tool$workspaceCurrent);
+  tool(tool$organizationsCurrent);
   tool(tool$tracesCreate);
+  tool(tool$tracesList);
   if (deps.dynamic) {
     registerDynamicTools(deps.logger, server, getClient, toolMap, scopes);
   }
@@ -63161,13 +63469,14 @@ var init_server2 = __esm(() => {
   init_evaluatorsCreateVersion2();
   init_evaluatorsGet2();
   init_evaluatorsList2();
+  init_organizationsCurrent2();
   init_testsBulkCreate2();
   init_testsCreate2();
   init_testsGet2();
   init_testsList2();
   init_testsUpdate2();
   init_tracesCreate2();
-  init_workspaceCurrent2();
+  init_tracesList2();
 });
 
 // src/tool-names.ts
@@ -63464,6 +63773,17 @@ the conversation history as input, what the agent produced as output,
 and optional metadata. Use this to capture live traffic for later
 review and conversion into tests. Fails with 429 once the workspace
 hits its trace limit.
+`
+    },
+    {
+      name: "list-traces",
+      description: `List the traces in the caller's org, newest first, one page at a
+time. Each trace carries a short preview of its last user message
+and reply, its tool calls and labels, and the scores from its
+latest scoring run with each evaluator's reasoning and evaluator
+version. Filter by agent, label, creation time, text, or score.
+Set include_content to also get each trace's full input, output
+and metadata.
 `
     }
   ];
@@ -64372,7 +64692,7 @@ 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/v0.0.87/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/v0.0.89/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>
@@ -67326,7 +67646,7 @@ var routes = buildRouteMap({
 var app = buildApplication(routes, {
   name: "mcp",
   versionInfo: {
-    currentVersion: "0.0.87"
+    currentVersion: "0.0.89"
   }
 });
 run(app, process3.argv.slice(2), buildContext(process3));
@@ -67334,5 +67654,5 @@ export {
   app
 };
 
-//# debugId=0E09DA234440D3F664756E2164756E21
+//# debugId=5A80FFDF0EAFC95564756E2164756E21
 //# sourceMappingURL=mcp-server.js.map