@tangle-network/agent-eval 1.0.11 → 1.0.12

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 CHANGED
@@ -4,6 +4,33 @@ All notable changes to `@tangle-network/agent-eval` and its sibling `agent-eval-
4
4
 
5
5
  ---
6
6
 
7
+ ## [1.0.12] — 2026-10-31
8
+
9
+ - fix(statistics): exact retired-contract compat for summarizeDistribution
10
+ - chore(release): 1.0.11
11
+ - fix(statistics): export summarizeDistribution from the package root
12
+ - chore(release): 1.0.10
13
+ - chore(release): 1.0.9
14
+ - feat(statistics): summarizeDistribution + percentileNearestRank (consolidation home)
15
+ - chore(release): 1.0.8
16
+ - chore(analyst): re-pin the implementation manifest on merged main
17
+ - feat(statistics,harness): two-sample Welch evidence + lifted sandbox assertion library
18
+ - chore(release): 1.0.7 (#1172)
19
+ - docs(statistics): keep inference policy and exports readable (#1173)
20
+ - feat(statistics): own independent effect summaries and ROC area (#1171)
21
+ - chore(release): 1.0.6 (#1170)
22
+ - refactor(reporting): share text table layout and simplify summaries (#1169)
23
+ - refactor(harness): retain unavailable verdicts through cleanup (#1168)
24
+ - chore(release): 1.0.5 (#1167)
25
+ - refactor(harness): normalize compact named expectations (#1166)
26
+ - chore(release): 1.0.4 (#1165)
27
+ - refactor(harness): own resource acquisition and readiness cleanup (#1164)
28
+ - chore(release): 1.0.3 (#1163)
29
+ - chore(release): declare the existing deployment runner for workflow lint
30
+ - fix(release): bind bootstrap to the existing qualified action revision
31
+ - fix(release): use the qualified pnpm bootstrap in publication workflows
32
+ - Refactor sandbox integration scenarios into the eval harness (#1161)
33
+
7
34
  ## [1.0.11] — 2026-10-30
8
35
 
9
36
  - fix(statistics): export summarizeDistribution from the package root
@@ -11,7 +11,7 @@ import { t as createDspyRlmTraceEngine } from "../dspy-rlm-engine-Xt3znknf.js";
11
11
  import { n as diffFindings, r as createChatTraceEngine, t as FindingsStore } from "../findings-store-Bp-0Guuv.js";
12
12
  import { a as systemOneAnalyst, p as validDecisionIdentity } from "../decision-point-CErE7GnP.js";
13
13
  import { t as createChatClient } from "../chat-client-D0me8iwq.js";
14
- import { A as agentRxBenchmarkCase, C as ANALYST_BENCHMARK_COST_LEDGER_FILE, D as AGENT_RX_UPSTREAM_REVISION, E as ANALYST_BENCHMARK_OBSERVATIONS_FILE, F as runAnalystBenchmark, I as traceStoreEvidenceResolver, M as normalizeAgentRxCategory, N as externalId, O as codeTraceBenchCase, P as positiveStep, S as ANALYST_BENCHMARK_IMPLEMENTATION_FILES, T as ANALYST_BENCHMARK_MANIFEST_FILE, _ as readAnalystBenchmarkArtifact, a as runInlineAnalystDefinition, b as ANALYST_BENCHMARK_EVIDENCE_DEPENDENCY_LOCK_SHA256, c as publicRlmAnalystDefinition, d as publicDirectAnalystDefinition, f as runChunkedAnalystDefinition, g as expandCodeTraceFailureBlocks, h as analystDefinitionProtocolSha256, i as primeCodeTraceAnalystDefinition, j as agentRxPredictionsToFindings, k as codeTracerPredictionsToFindings, l as runReplVariableAnalystDefinition, m as analystDefinitionAsymmetries, n as createPrimeBenchmarkRunner, o as loadPublicBenchmarkRows, p as AnalystExpressivenessError, r as primeAnalystProtocolSha256, s as createPublicBenchmarkRlmRunner, t as runAnalystBenchmarkCommand, u as createPublicBenchmarkDirectRunner, v as CODE_TRACE_BENCH_ANALYST_PROMPT, w as ANALYST_BENCHMARK_LOCAL_RECEIPT_FILE, x as ANALYST_BENCHMARK_EVIDENCE_IMPLEMENTATION_SHA256, y as publicBenchmarkProtocolSha256 } from "../benchmark-command-Did8Of2O.js";
14
+ import { A as agentRxBenchmarkCase, C as ANALYST_BENCHMARK_COST_LEDGER_FILE, D as AGENT_RX_UPSTREAM_REVISION, E as ANALYST_BENCHMARK_OBSERVATIONS_FILE, F as runAnalystBenchmark, I as traceStoreEvidenceResolver, M as normalizeAgentRxCategory, N as externalId, O as codeTraceBenchCase, P as positiveStep, S as ANALYST_BENCHMARK_IMPLEMENTATION_FILES, T as ANALYST_BENCHMARK_MANIFEST_FILE, _ as readAnalystBenchmarkArtifact, a as runInlineAnalystDefinition, b as ANALYST_BENCHMARK_EVIDENCE_DEPENDENCY_LOCK_SHA256, c as publicRlmAnalystDefinition, d as publicDirectAnalystDefinition, f as runChunkedAnalystDefinition, g as expandCodeTraceFailureBlocks, h as analystDefinitionProtocolSha256, i as primeCodeTraceAnalystDefinition, j as agentRxPredictionsToFindings, k as codeTracerPredictionsToFindings, l as runReplVariableAnalystDefinition, m as analystDefinitionAsymmetries, n as createPrimeBenchmarkRunner, o as loadPublicBenchmarkRows, p as AnalystExpressivenessError, r as primeAnalystProtocolSha256, s as createPublicBenchmarkRlmRunner, t as runAnalystBenchmarkCommand, u as createPublicBenchmarkDirectRunner, v as CODE_TRACE_BENCH_ANALYST_PROMPT, w as ANALYST_BENCHMARK_LOCAL_RECEIPT_FILE, x as ANALYST_BENCHMARK_EVIDENCE_IMPLEMENTATION_SHA256, y as publicBenchmarkProtocolSha256 } from "../benchmark-command-DZQrxqUL.js";
15
15
  import { a as normalizePrimeUsage, c as runPrimeExchange, i as mergePrimeRawUsage, n as buildPrimePrompt, o as primeProtocolSha256, r as buildPrimeRepairPrompt, s as projectPrimeTrajectory, t as analystUsageReceiptFromPrimeUsage } from "../prime-protocol-CrhpXRz9.js";
16
16
  import { t as nodeHttpPrimeBridgeTransport } from "../prime-bridge-transport-Csd5SQg7.js";
17
17
  import { t as clusterBootstrapMean } from "../cluster-bootstrap-hNXESoc1.js";
@@ -2021,7 +2021,7 @@ const ANALYST_BENCHMARK_IMPLEMENTATION_FILES = Object.freeze([
2021
2021
  "src/trace/redaction/detectors.ts",
2022
2022
  "src/verdict-cache.ts"
2023
2023
  ]);
2024
- const ANALYST_BENCHMARK_IMPLEMENTATION_SHA256 = "b480a601f52b64b4c949c92ecdf88fe22424c7c151b228e44dfc446eb3cc3019";
2024
+ const ANALYST_BENCHMARK_IMPLEMENTATION_SHA256 = "72c4aededb7ea91a1d8f7f9d5d8bf3794e7f9981338597ae00a99f8fa9a6e9ac";
2025
2025
  // src/analyst/benchmark-evidence-validation.ts
2026
2026
  const MIN_ACTION_EXCERPT_CHARACTERS = 12;
2027
2027
  const MAX_LABEL_SCAN_DEPTH = 64;
@@ -6858,4 +6858,4 @@ function shellQuote(value) {
6858
6858
  }
6859
6859
  export { agentRxBenchmarkCase as A, ANALYST_BENCHMARK_COST_LEDGER_FILE as C, AGENT_RX_UPSTREAM_REVISION as D, ANALYST_BENCHMARK_OBSERVATIONS_FILE as E, runAnalystBenchmark as F, traceStoreEvidenceResolver as I, normalizeAgentRxCategory as M, externalId as N, codeTraceBenchCase as O, positiveStep as P, ANALYST_BENCHMARK_IMPLEMENTATION_FILES as S, ANALYST_BENCHMARK_MANIFEST_FILE as T, readAnalystBenchmarkArtifact as _, runInlineAnalystDefinition as a, ANALYST_BENCHMARK_EVIDENCE_DEPENDENCY_LOCK_SHA256 as b, publicRlmAnalystDefinition as c, publicDirectAnalystDefinition as d, runChunkedAnalystDefinition as f, expandCodeTraceFailureBlocks as g, analystDefinitionProtocolSha256 as h, primeCodeTraceAnalystDefinition as i, agentRxPredictionsToFindings as j, codeTracerPredictionsToFindings as k, runReplVariableAnalystDefinition as l, analystDefinitionAsymmetries as m, createPrimeBenchmarkRunner as n, loadPublicBenchmarkRows as o, AnalystExpressivenessError as p, primeAnalystProtocolSha256 as r, createPublicBenchmarkRlmRunner as s, runAnalystBenchmarkCommand as t, createPublicBenchmarkDirectRunner as u, CODE_TRACE_BENCH_ANALYST_PROMPT as v, ANALYST_BENCHMARK_LOCAL_RECEIPT_FILE as w, ANALYST_BENCHMARK_EVIDENCE_IMPLEMENTATION_SHA256 as x, publicBenchmarkProtocolSha256 as y };
6860
6860
 
6861
- //# sourceMappingURL=benchmark-command-Did8Of2O.js.map
6861
+ //# sourceMappingURL=benchmark-command-DZQrxqUL.js.map