@tangle-network/agent-eval 0.221.0 → 0.221.2
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/CHANGELOG.md +70 -0
- package/dist/analyst/index.js +3 -3
- package/dist/{benchmark-command-8XQfpUK3.js → benchmark-command-DiGqnGUt.js} +2 -2
- package/dist/{benchmark-command-8XQfpUK3.js.map → benchmark-command-DiGqnGUt.js.map} +1 -1
- package/dist/cli.js +1 -1
- package/dist/completion-verifier-YwoeujPl.js.map +1 -1
- package/dist/contract/index.js +1 -1
- package/dist/{decision-point-Co3u3W8k.js → decision-point-0nJubuoM.js} +8 -2
- package/dist/{decision-point-Co3u3W8k.js.map → decision-point-0nJubuoM.js.map} +1 -1
- package/dist/{default-registry-BbFCkqvB.js → default-registry-BZ5FyYl9.js} +2 -2
- package/dist/{default-registry-BbFCkqvB.js.map → default-registry-BZ5FyYl9.js.map} +1 -1
- package/dist/index.js +1 -1
- package/dist/openapi.json +1 -1
- package/dist/systemone-D3MrW63C.d.ts.map +1 -1
- package/dist/systemone.js +1 -1
- package/package.json +4 -4
package/dist/index.js
CHANGED
|
@@ -34,7 +34,7 @@ import { a as LlmCallError, i as stripFencedJson, n as isRecord, o as LlmRespons
|
|
|
34
34
|
import { r as selfImprove, t as defineAgentEval } from "./define-agent-eval-CJhECJNL.js";
|
|
35
35
|
import { n as equivalenceVerdict, t as certificationEvidenceDigest } from "./verdict-BGQmxQ8A.js";
|
|
36
36
|
import { c as createBoundedTraceAnalysisStore, p as DEFAULT_TRACE_ANALYST_BUDGETS, t as createTraceAnalyst } from "./kind-factory-Dl0PmNQL.js";
|
|
37
|
-
import { C as snapshotAnalystRun, S as readAnalystReview, b as assertUniqueFindingIds, c as FAILURE_MODE_KIND_SPEC, i as DEFAULT_TRACE_ANALYST_KINDS, n as AnalystRegistry, t as buildDefaultAnalystRegistry, v as analystFindingDigest, w as validateAnalystReviewDecisions, x as completedAnalystReviewQuality, y as analystRunDigest } from "./default-registry-
|
|
37
|
+
import { C as snapshotAnalystRun, S as readAnalystReview, b as assertUniqueFindingIds, c as FAILURE_MODE_KIND_SPEC, i as DEFAULT_TRACE_ANALYST_KINDS, n as AnalystRegistry, t as buildDefaultAnalystRegistry, v as analystFindingDigest, w as validateAnalystReviewDecisions, x as completedAnalystReviewQuality, y as analystRunDigest } from "./default-registry-BZ5FyYl9.js";
|
|
38
38
|
import { OUTPUT_VALUE } from "./trace-attributes.js";
|
|
39
39
|
import { n as extractUsageFromSse, t as extractUsage } from "./extract-usage-BZLU01kR.js";
|
|
40
40
|
import { n as InMemoryRawProviderSink, r as NoopRawProviderSink, t as FileSystemRawProviderSink } from "./raw-provider-sink-75w4FWBj.js";
|
package/dist/openapi.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
"openapi": "3.1.0",
|
|
3
3
|
"info": {
|
|
4
4
|
"title": "@tangle-network/agent-eval — wire protocol",
|
|
5
|
-
"version": "0.221.
|
|
5
|
+
"version": "0.221.2",
|
|
6
6
|
"description": "HTTP and stdio RPC interface to agent-eval. The TypeScript runtime is the source of truth; this spec is the contract that cross-language clients (Python, Rust, Go) generate from.\n\nWire-protocol version: 1.1.0. Bumps on breaking changes to request/response schemas.",
|
|
7
7
|
"contact": {
|
|
8
8
|
"name": "Tangle Network",
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"systemone-D3MrW63C.d.ts","names":["_$__meta_eval0.calibrationFromPairs"],"sources":["../src/analyst/findings-store.ts","../src/analyst/findings-ledger.ts","../src/decision-point-records.ts","../src/decision-point.ts","../src/systemone-domain-match.ts","../src/systemone.ts"],"mappings":";;;;;;;;;;;;;;;UA8BiB,yBAAyB;EACxC;;cAGW;WAGiB;mBAFX;EAEjB,YAA4B;EAItB,OAAO,eAAe,UAAU,mBAAmB;;EAQzD,WAAW;;EAkBX,QAAQ,gBAAgB;;UAOT;;EAEf,UAAU;;EAEV,aAAa;;EAEb,WAAW;;;;;EAKX,SAAS;IAAQ,UAAU;IAAkB,SAAS;;;UAGvC;;;;;;;;EAQf,cAAc,UAAU,gBAAgB,SAAS;;;;;;iBAkBnC,aACd,UAAU,oBACV,SAAS,oBACT,SAAQ,aACP;;;;UC9Dc,sBAAsB;;EAErC;EACA,WAAW;;UAGI;EACf,UAAU;;EAEV,YAAY;;KAGF,sBAAsB;EAC5B;EAAiB,OAAO;;EAExB;EACA;IAAS;IAA8C;;;UAG5C;WACN;EACT,OACE,eACA,mBAAmB,mBAClB,QAAQ;IAAwB;IAAkB;;EACrD,QAAQ,QAAQ,sBAAsB;;;iBAwFxB,mBAAmB;EAAW;IAAiB;UA0C9C;EACf,QAAQ,wBAAwB;EAChC;EACA;EACA;EACA;;;;;EAKA;;;iBAIc,cAAc,UAAU;EAAmB;GACzD,eAAe,KACf,OAAO,gBACN;UAoBc;EACf;EACA;EACA,MAAM;;;;;;;iBAQQ,qBACd,eAAe,iBACf,SAAQ,aACP;;;cCpOU,qBAAmB,EAAA,uBAAA,EAAA;;;;;;;GAI9B,EAAA,KAAA;;KAGU,gBAAgB,EAAE,aAAa;cA4BrC,sBAAoB,EAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAiBf,EAAA,KAAA;KAEC,iBAAiB,EAAE,aAAa;;;KC7DhC,oBAAoB,eAAe,QAAQ;EAAqB;;UAC3D;;EAEf;EACA;EACA;EACA,SAAS,eAAe;;EAExB,cAAc;IACZ;IACA;IACA,cAAc;;;UAGD,qBAAqB,UAAU;EAC9C;EACA;EACA,QAAQ;EACR;EACA,WAAW;EACX,WAAW,WAAW,IAAI;EAC1B,QAAQ;EACR,WAAW;;UAEI;EACf;EACA;EACA,OAAO;EACP;EACA,OAAO;EACP,OAAO;EACP;EACA,cAAc;;UAEC;EACf,QAAQ;EACR,SAAS;EACT;;EAEA,QAAQ;;;iBAIM,uBAAuB;EACrC;EACA;EACA;EACA;EACA;EACA,UAAU;;;cAqBC;;iBAwGG,oBAAoB,UAAU,mBAAmB,SAAS,qBAAqB;;;
|
|
1
|
+
{"version":3,"file":"systemone-D3MrW63C.d.ts","names":["_$__meta_eval0.calibrationFromPairs"],"sources":["../src/analyst/findings-store.ts","../src/analyst/findings-ledger.ts","../src/decision-point-records.ts","../src/decision-point.ts","../src/systemone-domain-match.ts","../src/systemone.ts"],"mappings":";;;;;;;;;;;;;;;UA8BiB,yBAAyB;EACxC;;cAGW;WAGiB;mBAFX;EAEjB,YAA4B;EAItB,OAAO,eAAe,UAAU,mBAAmB;;EAQzD,WAAW;;EAkBX,QAAQ,gBAAgB;;UAOT;;EAEf,UAAU;;EAEV,aAAa;;EAEb,WAAW;;;;;EAKX,SAAS;IAAQ,UAAU;IAAkB,SAAS;;;UAGvC;;;;;;;;EAQf,cAAc,UAAU,gBAAgB,SAAS;;;;;;iBAkBnC,aACd,UAAU,oBACV,SAAS,oBACT,SAAQ,aACP;;;;UC9Dc,sBAAsB;;EAErC;EACA,WAAW;;UAGI;EACf,UAAU;;EAEV,YAAY;;KAGF,sBAAsB;EAC5B;EAAiB,OAAO;;EAExB;EACA;IAAS;IAA8C;;;UAG5C;WACN;EACT,OACE,eACA,mBAAmB,mBAClB,QAAQ;IAAwB;IAAkB;;EACrD,QAAQ,QAAQ,sBAAsB;;;iBAwFxB,mBAAmB;EAAW;IAAiB;UA0C9C;EACf,QAAQ,wBAAwB;EAChC;EACA;EACA;EACA;;;;;EAKA;;;iBAIc,cAAc,UAAU;EAAmB;GACzD,eAAe,KACf,OAAO,gBACN;UAoBc;EACf;EACA;EACA,MAAM;;;;;;;iBAQQ,qBACd,eAAe,iBACf,SAAQ,aACP;;;cCpOU,qBAAmB,EAAA,uBAAA,EAAA;;;;;;;GAI9B,EAAA,KAAA;;KAGU,gBAAgB,EAAE,aAAa;cA4BrC,sBAAoB,EAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAiBf,EAAA,KAAA;KAEC,iBAAiB,EAAE,aAAa;;;KC7DhC,oBAAoB,eAAe,QAAQ;EAAqB;;UAC3D;;EAEf;EACA;EACA;EACA,SAAS,eAAe;;EAExB,cAAc;IACZ;IACA;IACA,cAAc;;;UAGD,qBAAqB,UAAU;EAC9C;EACA;EACA,QAAQ;EACR;EACA,WAAW;EACX,WAAW,WAAW,IAAI;EAC1B,QAAQ;EACR,WAAW;;UAEI;EACf;EACA;EACA,OAAO;EACP;EACA,OAAO;EACP,OAAO;EACP;EACA,cAAc;;UAEC;EACf,QAAQ;EACR,SAAS;EACT;;EAEA,QAAQ;;;iBAIM,uBAAuB;EACrC;EACA;EACA;EACA;EACA;EACA,UAAU;;;cAqBC;;iBAwGG,oBAAoB,UAAU,mBAAmB,SAAS,qBAAqB;;;EAyE7E,OAAA,YAAA,oBAAkB,UACrB,oBACR,QAAQ,sBAAsB;EAmHZ,QAAA;IACnB;IACA;IACA,QAAQ;IACR;IACA,cAAc;MACZ,QAAQ;IAAwB;IAAkB;;;;;;;;;;;;;;;+BAAlDA;;;;;;;;;;;KCjYI;UAEK;GACd,cAAc;EACf;EACA;;UAGe;EACf;EACA;EACA;EACA;EACA;;UAGe;EACf;EACA;EACA;EACA,YAAY;EACZ,UAAU;EACV,UAAU;EACV,YAAY;;KAGT;EACH;IACE;IACA;IACA;MAAY;MAAe;MAAkB;;;EAE/C;IACE;IACA;IACA,UAAU;;EAEZ;IACE;IACA;IACA,UAAU;;;UAIG,qCAAqC;EACpD,SAAS,iBAAiB;;EAE1B;;KAGU;UAUK;EACf;EACA;EACA;EACA;EACA;EACA,UAAU;EACV,QAAQ;EACR;IACE,OAAO;IACP,UAAU;IACV,QAAQ;;EAEV;IACE;IACA;IACA;;EAEF;;;iBAqHc,4BACd,OAAO,4BACN;;iBA4Ba,sBACd,eAAe,wBACf;;iBAwBc,2BACd,UAAU,8BACV,eACC;;;KCtNS,oBAAoB,iBAAiB;UAChC;EACf,UAAU;EACV,aAAa,iBAAiB;EAC9B,gBAAgB,iBAAiB;EACjC,UAAU;EACV,WAAW,sBAAsB;EACjC,oBAAoB,OAAO,UAAU;;EAErC,eAAe,mBAAmB;;;iBAIpB,mBAAmB,SAAS,mCA2BtB,UAAU,oBAAkB,SACrC,iBAAiB,IAAE,UAClB,sBACT,QAAQ,iBAAiB,gBAAgB;KAUzC,MAAM,GAAG,UAAU;EAAc,UAAU;EAAG,UAAU;;KACxD,eAAe,GAAG,UAAU,sBAAsB,MAAM,OAAO,MAAM,IAAI,QAAQ;UACrE,yBAAyB;EACxC;EACA;EACA,WAAW;;UAEI,sBACf,GACA,UAAU,WAAW,UACrB,UAAU,qBAAqB,4BACvB;EACR;EACA;EACA,WAAW,eAAe,MAAM,GAAG,IAAI;EACvC,cACE,OAAO,MAAM,GAAG,IAChB,SAAS,sBACN,iBAAiB,QAAQ;EAC9B,aAAa,YAAY,GAAG;EAC5B,UAAU;EACV,OAAO,OAAO,gBAAgB,IAAI,OAAO,MAAM,GAAG,OAAO,aAAa,QAAQ;EAC9E,UACE,QAAQ,iBAAiB,gBAAgB,KACzC,OAAO,MAAM,GAAG,cACN;EACZ,aAAa,UAAU;;;iBA0BT,yBACd,QAAQ,iBACR,UAAU,yBACT;;iBAgCa,eACd,GACA,UAAU,WAAW,UACrB,UAAU,qBAAqB,oBAC/B,cAAc,SAAS,sBAAsB,GAAG,GAAG,KAAK,YAAY,GAAG;UAqDxD,wBAAwB,GAAG,UAAU,qBAAqB,4BACjE;EACR;EACA;EACA,WAAW,QAAQ;EACnB;EACA;EACA,WAAW,MAAM,OAAO,GAAG,SAAS,mBAAmB,IAAI,QAAQ;EACnE,cAAc,OAAO,GAAG,SAAS,mBAAmB,iBAAiB,QAAQ;EAC7E,WACE,OAAO,gBAAgB,IACvB,OAAO,GACP,SAAS,mBACN,mBAAmB,QAAQ;EAChC,UACE,QAAQ,iBAAiB,gBAAgB,KACzC,OAAO,GACP,SAAS,0BACC;;iBAGE,iBAAiB,GAAG,UAAU,qBAAqB,oBACjE,SAAS,wBAAwB,GAAG,KACnC,QAAQ"}
|
package/dist/systemone.js
CHANGED
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
import { SystemOneResponseError, normalizeSystemOneResult, parseSystemOneRecordedRequest, parseSystemOneRequest, parseSystemOneResult, systemOneConfidence, systemOneUsage } from "./systemone-protocol.js";
|
|
2
|
-
import { a as systemOneAnalyst, c as assessSystemOneDomainMatch, i as normalizedSystemOneScore, l as prepareSystemOneDomainMatch, n as decisionPoint, o as systemOneEvaluator, r as decisionQuestionDigest, s as systemOneJudge, t as DECISION_POINT_MIN_KAPPA, u as systemOneChoiceMargin } from "./decision-point-
|
|
2
|
+
import { a as systemOneAnalyst, c as assessSystemOneDomainMatch, i as normalizedSystemOneScore, l as prepareSystemOneDomainMatch, n as decisionPoint, o as systemOneEvaluator, r as decisionQuestionDigest, s as systemOneJudge, t as DECISION_POINT_MIN_KAPPA, u as systemOneChoiceMargin } from "./decision-point-0nJubuoM.js";
|
|
3
3
|
import { n as asJudge, r as createEvaluator, t as asAnalyst } from "./evaluation-wdGyZ7jV.js";
|
|
4
4
|
import { n as prepareSystemOneReview, r as systemOneReviewFindings, t as assessSystemOneReview } from "./systemone-review-DFGtIUcf.js";
|
|
5
5
|
export { DECISION_POINT_MIN_KAPPA, SystemOneResponseError, asAnalyst, asJudge, assessSystemOneDomainMatch, assessSystemOneReview, createEvaluator, decisionPoint, decisionQuestionDigest, normalizeSystemOneResult, normalizedSystemOneScore, parseSystemOneRecordedRequest, parseSystemOneRequest, parseSystemOneResult, prepareSystemOneDomainMatch, prepareSystemOneReview, systemOneAnalyst, systemOneChoiceMargin, systemOneConfidence, systemOneEvaluator, systemOneJudge, systemOneReviewFindings, systemOneUsage };
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@tangle-network/agent-eval",
|
|
3
|
-
"version": "0.221.
|
|
3
|
+
"version": "0.221.2",
|
|
4
4
|
"description": "Evaluate and improve AI agents from runs, traces, judges, and feedback. Compare candidates, cluster failures, measure lift, and gate releases.",
|
|
5
5
|
"homepage": "https://github.com/tangle-network/agent-eval#readme",
|
|
6
6
|
"repository": {
|
|
@@ -229,15 +229,15 @@
|
|
|
229
229
|
"dependencies": {
|
|
230
230
|
"@asteasolutions/zod-to-openapi": "^9.1.0",
|
|
231
231
|
"@hono/node-server": "^2.1.1",
|
|
232
|
-
"@tangle-network/agent-core": "^0.13.
|
|
233
|
-
"@tangle-network/agent-interface": "^5.
|
|
232
|
+
"@tangle-network/agent-core": "^0.13.2",
|
|
233
|
+
"@tangle-network/agent-interface": "^5.7.1",
|
|
234
234
|
"@tangle-network/agent-trace-contract": "^2.0.0",
|
|
235
235
|
"@tangle-network/harness-sessions": ">=0.1.0 <0.2.0",
|
|
236
236
|
"hono": "^4.13.9",
|
|
237
237
|
"linear-sum-assignment": "1.0.9",
|
|
238
238
|
"re2js": "2.8.6",
|
|
239
239
|
"smol-toml": "1.9.0",
|
|
240
|
-
"zod": "4.6.
|
|
240
|
+
"zod": "4.6.6"
|
|
241
241
|
},
|
|
242
242
|
"devDependencies": {
|
|
243
243
|
"@arethetypeswrong/cli": "^0.18.5",
|