@dalmia/calibrate-mcp 0.0.43 → 0.0.44
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/bin/mcp-server.js +19 -7
- package/bin/mcp-server.js.map +9 -8
- package/esm/landing-page.js +1 -1
- package/esm/lib/config.d.ts +2 -2
- package/esm/lib/config.js +2 -2
- package/esm/mcp-server/mcp-server.js +1 -1
- package/esm/mcp-server/server.js +1 -1
- package/esm/models/agenttestrunlistitem.d.ts +2 -1
- package/esm/models/agenttestrunlistitem.d.ts.map +1 -1
- package/esm/models/agenttestrunlistitem.js +2 -1
- package/esm/models/agenttestrunlistitem.js.map +1 -1
- package/esm/models/runlistevaluator.d.ts +13 -0
- package/esm/models/runlistevaluator.d.ts.map +1 -0
- package/esm/models/runlistevaluator.js +9 -0
- package/esm/models/runlistevaluator.js.map +1 -0
- package/package.json +1 -1
- package/src/landing-page.ts +1 -1
- package/src/lib/config.ts +2 -2
- package/src/mcp-server/mcp-server.ts +1 -1
- package/src/mcp-server/server.ts +1 -1
- package/src/models/agenttestrunlistitem.ts +7 -3
- package/src/models/runlistevaluator.ts +27 -0
package/esm/landing-page.js
CHANGED
|
@@ -915,7 +915,7 @@ http_headers = { "api-key-auth" = "YOUR_API_KEY_AUTH" }`;
|
|
|
915
915
|
<h1>Instructions</h1>
|
|
916
916
|
<p>One-click installation for Claude Desktop users</p>
|
|
917
917
|
<div class="instruction-item">
|
|
918
|
-
<a href="https://github.com/dalmia/calibrate-mcp/releases/download/v0.0.
|
|
918
|
+
<a href="https://github.com/dalmia/calibrate-mcp/releases/download/v0.0.44/mcp-server.mcpb" download="mcp-server.mcpb" class="action-button header-action" style="display: inline-flex; margin-bottom: 16px;">
|
|
919
919
|
📥 Download MCP Bundle
|
|
920
920
|
</a>
|
|
921
921
|
</div>
|
package/esm/lib/config.d.ts
CHANGED
|
@@ -35,8 +35,8 @@ export declare function serverURLFromOptions(options: SDKOptions): URL | null;
|
|
|
35
35
|
export declare const SDK_METADATA: {
|
|
36
36
|
readonly language: "typescript";
|
|
37
37
|
readonly openapiDocVersion: "0.1.0";
|
|
38
|
-
readonly sdkVersion: "0.0.
|
|
38
|
+
readonly sdkVersion: "0.0.44";
|
|
39
39
|
readonly genVersion: "2.915.1";
|
|
40
|
-
readonly userAgent: "speakeasy-sdk/mcp-typescript 0.0.
|
|
40
|
+
readonly userAgent: "speakeasy-sdk/mcp-typescript 0.0.44 2.915.1 0.1.0 @dalmia/calibrate-mcp";
|
|
41
41
|
};
|
|
42
42
|
//# sourceMappingURL=config.d.ts.map
|
package/esm/lib/config.js
CHANGED
|
@@ -27,8 +27,8 @@ export function serverURLFromOptions(options) {
|
|
|
27
27
|
export const SDK_METADATA = {
|
|
28
28
|
language: "typescript",
|
|
29
29
|
openapiDocVersion: "0.1.0",
|
|
30
|
-
sdkVersion: "0.0.
|
|
30
|
+
sdkVersion: "0.0.44",
|
|
31
31
|
genVersion: "2.915.1",
|
|
32
|
-
userAgent: "speakeasy-sdk/mcp-typescript 0.0.
|
|
32
|
+
userAgent: "speakeasy-sdk/mcp-typescript 0.0.44 2.915.1 0.1.0 @dalmia/calibrate-mcp",
|
|
33
33
|
};
|
|
34
34
|
//# sourceMappingURL=config.js.map
|
package/esm/mcp-server/server.js
CHANGED
|
@@ -49,7 +49,7 @@ import { tool$tracesCreate } from "./tools/tracesCreate.js";
|
|
|
49
49
|
export function createMCPServer(deps) {
|
|
50
50
|
const server = new McpServer({
|
|
51
51
|
name: "CalibrateMcp",
|
|
52
|
-
version: "0.0.
|
|
52
|
+
version: "0.0.44",
|
|
53
53
|
});
|
|
54
54
|
const getClient = deps.getSDK || (() => new CalibrateMcpCore({
|
|
55
55
|
security: deps.security,
|
|
@@ -1,6 +1,7 @@
|
|
|
1
1
|
import * as z from "zod";
|
|
2
2
|
import { ClosedEnum } from "../types/enums.js";
|
|
3
3
|
import { ModelRunSummary } from "./modelrunsummary.js";
|
|
4
|
+
import { RunListEvaluator } from "./runlistevaluator.js";
|
|
4
5
|
import { TaskStatus } from "./taskstatus.js";
|
|
5
6
|
import { TestRunCaseSummary } from "./testruncasesummary.js";
|
|
6
7
|
/**
|
|
@@ -36,7 +37,7 @@ export type AgentTestRunListItem = {
|
|
|
36
37
|
total_tests?: number | null | undefined;
|
|
37
38
|
passed?: number | null | undefined;
|
|
38
39
|
failed?: number | null | undefined;
|
|
39
|
-
evaluators?: Array<
|
|
40
|
+
evaluators?: Array<RunListEvaluator> | undefined;
|
|
40
41
|
results?: Array<TestRunCaseSummary> | null | undefined;
|
|
41
42
|
latency_ms?: {
|
|
42
43
|
[k: string]: any;
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"agenttestrunlistitem.d.ts","sourceRoot":"","sources":["../../src/models/agenttestrunlistitem.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AACzB,OAAO,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AAC/C,OAAO,EACL,eAAe,EAEhB,MAAM,sBAAsB,CAAC;AAC9B,OAAO,EAAE,UAAU,EAAwB,MAAM,iBAAiB,CAAC;AACnE,OAAO,EACL,kBAAkB,EAEnB,MAAM,yBAAyB,CAAC;AAEjC;;;;;;GAMG;AACH,eAAO,MAAM,wBAAwB;;;CAG3B,CAAC;AACX;;;;;;GAMG;AACH,MAAM,MAAM,wBAAwB,GAAG,UAAU,CAC/C,OAAO,wBAAwB,CAChC,CAAC;AAEF,eAAO,MAAM,kCAAkC;;;EAK9C,CAAC;AAEF,MAAM,MAAM,oBAAoB,GAAG;IACjC,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,UAAU,CAAC;IACnB,IAAI,EAAE,wBAAwB,CAAC;IAC/B,UAAU,EAAE,MAAM,CAAC;IACnB,UAAU,EAAE,MAAM,CAAC;IACnB,WAAW,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;IACxC,MAAM,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;IACnC,MAAM,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;IACnC,UAAU,CAAC,EAAE,KAAK,CAAC,
|
|
1
|
+
{"version":3,"file":"agenttestrunlistitem.d.ts","sourceRoot":"","sources":["../../src/models/agenttestrunlistitem.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AACzB,OAAO,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AAC/C,OAAO,EACL,eAAe,EAEhB,MAAM,sBAAsB,CAAC;AAC9B,OAAO,EACL,gBAAgB,EAEjB,MAAM,uBAAuB,CAAC;AAC/B,OAAO,EAAE,UAAU,EAAwB,MAAM,iBAAiB,CAAC;AACnE,OAAO,EACL,kBAAkB,EAEnB,MAAM,yBAAyB,CAAC;AAEjC;;;;;;GAMG;AACH,eAAO,MAAM,wBAAwB;;;CAG3B,CAAC;AACX;;;;;;GAMG;AACH,MAAM,MAAM,wBAAwB,GAAG,UAAU,CAC/C,OAAO,wBAAwB,CAChC,CAAC;AAEF,eAAO,MAAM,kCAAkC;;;EAK9C,CAAC;AAEF,MAAM,MAAM,oBAAoB,GAAG;IACjC,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,UAAU,CAAC;IACnB,IAAI,EAAE,wBAAwB,CAAC;IAC/B,UAAU,EAAE,MAAM,CAAC;IACnB,UAAU,EAAE,MAAM,CAAC;IACnB,WAAW,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;IACxC,MAAM,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;IACnC,MAAM,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;IACnC,UAAU,CAAC,EAAE,KAAK,CAAC,gBAAgB,CAAC,GAAG,SAAS,CAAC;IACjD,OAAO,CAAC,EAAE,KAAK,CAAC,kBAAkB,CAAC,GAAG,IAAI,GAAG,SAAS,CAAC;IACvD,UAAU,CAAC,EAAE;QAAE,CAAC,CAAC,EAAE,MAAM,GAAG,GAAG,CAAA;KAAE,GAAG,IAAI,GAAG,SAAS,CAAC;IACrD,IAAI,CAAC,EAAE;QAAE,CAAC,CAAC,EAAE,MAAM,GAAG,GAAG,CAAA;KAAE,GAAG,IAAI,GAAG,SAAS,CAAC;IAC/C,YAAY,CAAC,EAAE;QAAE,CAAC,CAAC,EAAE,MAAM,GAAG,GAAG,CAAA;KAAE,GAAG,IAAI,GAAG,SAAS,CAAC;IACvD,aAAa,CAAC,EAAE,KAAK,CAAC,eAAe,CAAC,GAAG,IAAI,GAAG,SAAS,CAAC;IAC1D,KAAK,CAAC,EAAE,OAAO,GAAG,SAAS,CAAC;IAC5B,SAAS,CAAC,EAAE,OAAO,GAAG,SAAS,CAAC;IAChC,WAAW,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;CACzC,CAAC;AAEF,eAAO,MAAM,8BAA8B,EAAE,CAAC,CAAC,OAAO,CAAC,oBAAoB,CAkDvE,CAAC"}
|
|
@@ -3,6 +3,7 @@
|
|
|
3
3
|
*/
|
|
4
4
|
import * as z from "zod";
|
|
5
5
|
import { ModelRunSummary$zodSchema, } from "./modelrunsummary.js";
|
|
6
|
+
import { RunListEvaluator$zodSchema, } from "./runlistevaluator.js";
|
|
6
7
|
import { TaskStatus$zodSchema } from "./taskstatus.js";
|
|
7
8
|
import { TestRunCaseSummary$zodSchema, } from "./testruncasesummary.js";
|
|
8
9
|
/**
|
|
@@ -25,7 +26,7 @@ export const AgentTestRunListItem$zodSchema = z
|
|
|
25
26
|
cost: z.record(z.string(), z.any()).nullable().optional().describe("Aggregated cost as `{mean, min, max, count}` (USD)"),
|
|
26
27
|
created_at: z.string().describe("When the run was created (ISO 8601 UTC)"),
|
|
27
28
|
error: z.boolean().default(false).describe("True if the run failed"),
|
|
28
|
-
evaluators: z.array(
|
|
29
|
+
evaluators: z.array(RunListEvaluator$zodSchema).optional().describe("The evaluators that judged this run, deduplicated and in display order. A `Tool call` entry is appended when any test in the run was a tool-call test. That entry has no `uuid`, because it is not an evaluator in the library. Empty when the run had no evaluators"),
|
|
29
30
|
failed: z.int().nullable().optional().describe("Number of test cases that failed"),
|
|
30
31
|
is_public: z.boolean().default(false).describe("Whether the run is shared publicly"),
|
|
31
32
|
latency_ms: z.record(z.string(), z.any()).nullable().optional().describe("Aggregated latency in milliseconds, as `{p50, p95, p99, count}`"),
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"agenttestrunlistitem.js","sourceRoot":"","sources":["../../src/models/agenttestrunlistitem.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AAEzB,OAAO,EAEL,yBAAyB,GAC1B,MAAM,sBAAsB,CAAC;AAC9B,OAAO,EAAc,oBAAoB,EAAE,MAAM,iBAAiB,CAAC;AACnE,OAAO,EAEL,4BAA4B,GAC7B,MAAM,yBAAyB,CAAC;AAEjC;;;;;;GAMG;AACH,MAAM,CAAC,MAAM,wBAAwB,GAAG;IACtC,WAAW,EAAE,eAAe;IAC5B,YAAY,EAAE,eAAe;CACrB,CAAC;AAYX,MAAM,CAAC,MAAM,kCAAkC,GAAG,CAAC,CAAC,IAAI,CAAC;IACvD,eAAe;IACf,eAAe;CAChB,CAAC,CAAC,QAAQ,CACT,8HAA8H,CAC/H,CAAC;AAuBF,MAAM,CAAC,MAAM,8BAA8B,GAAoC,CAAC;KAC7E,MAAM,CAAC;IACN,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC,GAAG,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAChE,oDAAoD,CACrD;IACD,UAAU,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CAAC,yCAAyC,CAAC;IAC1E,KAAK,EAAE,CAAC,CAAC,OAAO,EAAE,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,QAAQ,CAAC,wBAAwB,CAAC;IACpE,UAAU,EAAE,CAAC,CAAC,KAAK,CAAC,
|
|
1
|
+
{"version":3,"file":"agenttestrunlistitem.js","sourceRoot":"","sources":["../../src/models/agenttestrunlistitem.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AAEzB,OAAO,EAEL,yBAAyB,GAC1B,MAAM,sBAAsB,CAAC;AAC9B,OAAO,EAEL,0BAA0B,GAC3B,MAAM,uBAAuB,CAAC;AAC/B,OAAO,EAAc,oBAAoB,EAAE,MAAM,iBAAiB,CAAC;AACnE,OAAO,EAEL,4BAA4B,GAC7B,MAAM,yBAAyB,CAAC;AAEjC;;;;;;GAMG;AACH,MAAM,CAAC,MAAM,wBAAwB,GAAG;IACtC,WAAW,EAAE,eAAe;IAC5B,YAAY,EAAE,eAAe;CACrB,CAAC;AAYX,MAAM,CAAC,MAAM,kCAAkC,GAAG,CAAC,CAAC,IAAI,CAAC;IACvD,eAAe;IACf,eAAe;CAChB,CAAC,CAAC,QAAQ,CACT,8HAA8H,CAC/H,CAAC;AAuBF,MAAM,CAAC,MAAM,8BAA8B,GAAoC,CAAC;KAC7E,MAAM,CAAC;IACN,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC,GAAG,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAChE,oDAAoD,CACrD;IACD,UAAU,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CAAC,yCAAyC,CAAC;IAC1E,KAAK,EAAE,CAAC,CAAC,OAAO,EAAE,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,QAAQ,CAAC,wBAAwB,CAAC;IACpE,UAAU,EAAE,CAAC,CAAC,KAAK,CAAC,0BAA0B,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,CACjE,sQAAsQ,CACvQ;IACD,MAAM,EAAE,CAAC,CAAC,GAAG,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAC5C,kCAAkC,CACnC;IACD,SAAS,EAAE,CAAC,CAAC,OAAO,EAAE,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,QAAQ,CAC5C,oCAAoC,CACrC;IACD,UAAU,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC,GAAG,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CACtE,iEAAiE,CAClE;IACD,aAAa,EAAE,CAAC,CAAC,KAAK,CAAC,yBAAyB,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE;SACpE,QAAQ,CACP,8FAA8F,CAC/F;IACH,IAAI,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CACvB,gFAAgF,CACjF;IACD,MAAM,EAAE,CAAC,CAAC,GAAG,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAC5C,kCAAkC,CACnC;IACD,OAAO,EAAE,CAAC,CAAC,KAAK,CAAC,4BAA4B,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE;SACjE,QAAQ,CACP,mFAAmF,CACpF;IACH,WAAW,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CACpD,yCAAyC,CAC1C;IACD,MAAM,EAAE,oBAAoB;IAC5B,WAAW,EAAE,CAAC,CAAC,GAAG,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CACjD,4BAA4B,CAC7B;IACD,YAAY,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC,GAAG,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CACxE,qDAAqD,CACtD;IACD,IAAI,EAAE,kCAAkC,CAAC,QAAQ,CAC/C,8HAA8H,CAC/H;IACD,UAAU,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CAC7B,8CAA8C,CAC/C;IACD,IAAI,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CAAC,iBAAiB,CAAC;CAC7C,CAAC,CAAC"}
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
import * as z from "zod";
|
|
2
|
+
/**
|
|
3
|
+
* One evaluator on a run-LIST row: just enough to name it and open it.
|
|
4
|
+
*
|
|
5
|
+
* @remarks
|
|
6
|
+
* The rubric, prompt and version live on the run-DETAIL endpoint.
|
|
7
|
+
*/
|
|
8
|
+
export type RunListEvaluator = {
|
|
9
|
+
uuid?: string | null | undefined;
|
|
10
|
+
name: string;
|
|
11
|
+
};
|
|
12
|
+
export declare const RunListEvaluator$zodSchema: z.ZodType<RunListEvaluator>;
|
|
13
|
+
//# sourceMappingURL=runlistevaluator.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"runlistevaluator.d.ts","sourceRoot":"","sources":["../../src/models/runlistevaluator.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AAEzB;;;;;GAKG;AACH,MAAM,MAAM,gBAAgB,GAAG;IAC7B,IAAI,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,SAAS,CAAC;IACjC,IAAI,EAAE,MAAM,CAAC;CACd,CAAC;AAEF,eAAO,MAAM,0BAA0B,EAAE,CAAC,CAAC,OAAO,CAAC,gBAAgB,CASlE,CAAC"}
|
|
@@ -0,0 +1,9 @@
|
|
|
1
|
+
/*
|
|
2
|
+
* Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
|
|
3
|
+
*/
|
|
4
|
+
import * as z from "zod";
|
|
5
|
+
export const RunListEvaluator$zodSchema = z.object({
|
|
6
|
+
name: z.string().describe("Name of the evaluator"),
|
|
7
|
+
uuid: z.string().nullable().optional().describe("ID of the evaluator. Absent for the `Tool call` entry, which is not an evaluator in the library"),
|
|
8
|
+
}).describe("One evaluator on a run-LIST row: just enough to name it and open it.\nThe rubric, prompt and version live on the run-DETAIL endpoint.");
|
|
9
|
+
//# sourceMappingURL=runlistevaluator.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"runlistevaluator.js","sourceRoot":"","sources":["../../src/models/runlistevaluator.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,KAAK,CAAC,MAAM,KAAK,CAAC;AAazB,MAAM,CAAC,MAAM,0BAA0B,GAAgC,CAAC,CAAC,MAAM,CAC7E;IACE,IAAI,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CAAC,uBAAuB,CAAC;IAClD,IAAI,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAC7C,iGAAiG,CAClG;CACF,CACF,CAAC,QAAQ,CACR,uIAAuI,CACxI,CAAC"}
|
package/package.json
CHANGED
package/src/landing-page.ts
CHANGED
|
@@ -930,7 +930,7 @@ http_headers = { "api-key-auth" = "YOUR_API_KEY_AUTH" }`;
|
|
|
930
930
|
<h1>Instructions</h1>
|
|
931
931
|
<p>One-click installation for Claude Desktop users</p>
|
|
932
932
|
<div class="instruction-item">
|
|
933
|
-
<a href="https://github.com/dalmia/calibrate-mcp/releases/download/v0.0.
|
|
933
|
+
<a href="https://github.com/dalmia/calibrate-mcp/releases/download/v0.0.44/mcp-server.mcpb" download="mcp-server.mcpb" class="action-button header-action" style="display: inline-flex; margin-bottom: 16px;">
|
|
934
934
|
📥 Download MCP Bundle
|
|
935
935
|
</a>
|
|
936
936
|
</div>
|
package/src/lib/config.ts
CHANGED
|
@@ -65,8 +65,8 @@ export function serverURLFromOptions(options: SDKOptions): URL | null {
|
|
|
65
65
|
export const SDK_METADATA = {
|
|
66
66
|
language: "typescript",
|
|
67
67
|
openapiDocVersion: "0.1.0",
|
|
68
|
-
sdkVersion: "0.0.
|
|
68
|
+
sdkVersion: "0.0.44",
|
|
69
69
|
genVersion: "2.915.1",
|
|
70
70
|
userAgent:
|
|
71
|
-
"speakeasy-sdk/mcp-typescript 0.0.
|
|
71
|
+
"speakeasy-sdk/mcp-typescript 0.0.44 2.915.1 0.1.0 @dalmia/calibrate-mcp",
|
|
72
72
|
} as const;
|
package/src/mcp-server/server.ts
CHANGED
|
@@ -8,6 +8,10 @@ import {
|
|
|
8
8
|
ModelRunSummary,
|
|
9
9
|
ModelRunSummary$zodSchema,
|
|
10
10
|
} from "./modelrunsummary.js";
|
|
11
|
+
import {
|
|
12
|
+
RunListEvaluator,
|
|
13
|
+
RunListEvaluator$zodSchema,
|
|
14
|
+
} from "./runlistevaluator.js";
|
|
11
15
|
import { TaskStatus, TaskStatus$zodSchema } from "./taskstatus.js";
|
|
12
16
|
import {
|
|
13
17
|
TestRunCaseSummary,
|
|
@@ -53,7 +57,7 @@ export type AgentTestRunListItem = {
|
|
|
53
57
|
total_tests?: number | null | undefined;
|
|
54
58
|
passed?: number | null | undefined;
|
|
55
59
|
failed?: number | null | undefined;
|
|
56
|
-
evaluators?: Array<
|
|
60
|
+
evaluators?: Array<RunListEvaluator> | undefined;
|
|
57
61
|
results?: Array<TestRunCaseSummary> | null | undefined;
|
|
58
62
|
latency_ms?: { [k: string]: any } | null | undefined;
|
|
59
63
|
cost?: { [k: string]: any } | null | undefined;
|
|
@@ -71,8 +75,8 @@ export const AgentTestRunListItem$zodSchema: z.ZodType<AgentTestRunListItem> = z
|
|
|
71
75
|
),
|
|
72
76
|
created_at: z.string().describe("When the run was created (ISO 8601 UTC)"),
|
|
73
77
|
error: z.boolean().default(false).describe("True if the run failed"),
|
|
74
|
-
evaluators: z.array(
|
|
75
|
-
"
|
|
78
|
+
evaluators: z.array(RunListEvaluator$zodSchema).optional().describe(
|
|
79
|
+
"The evaluators that judged this run, deduplicated and in display order. A `Tool call` entry is appended when any test in the run was a tool-call test. That entry has no `uuid`, because it is not an evaluator in the library. Empty when the run had no evaluators",
|
|
76
80
|
),
|
|
77
81
|
failed: z.int().nullable().optional().describe(
|
|
78
82
|
"Number of test cases that failed",
|
|
@@ -0,0 +1,27 @@
|
|
|
1
|
+
/*
|
|
2
|
+
* Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
|
|
3
|
+
*/
|
|
4
|
+
|
|
5
|
+
import * as z from "zod";
|
|
6
|
+
|
|
7
|
+
/**
|
|
8
|
+
* One evaluator on a run-LIST row: just enough to name it and open it.
|
|
9
|
+
*
|
|
10
|
+
* @remarks
|
|
11
|
+
* The rubric, prompt and version live on the run-DETAIL endpoint.
|
|
12
|
+
*/
|
|
13
|
+
export type RunListEvaluator = {
|
|
14
|
+
uuid?: string | null | undefined;
|
|
15
|
+
name: string;
|
|
16
|
+
};
|
|
17
|
+
|
|
18
|
+
export const RunListEvaluator$zodSchema: z.ZodType<RunListEvaluator> = z.object(
|
|
19
|
+
{
|
|
20
|
+
name: z.string().describe("Name of the evaluator"),
|
|
21
|
+
uuid: z.string().nullable().optional().describe(
|
|
22
|
+
"ID of the evaluator. Absent for the `Tool call` entry, which is not an evaluator in the library",
|
|
23
|
+
),
|
|
24
|
+
},
|
|
25
|
+
).describe(
|
|
26
|
+
"One evaluator on a run-LIST row: just enough to name it and open it.\nThe rubric, prompt and version live on the run-DETAIL endpoint.",
|
|
27
|
+
);
|