@tangle-network/agent-eval 0.18.0 → 0.19.0

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.
Files changed (7) hide show
  1. package/README.md +1 -1
  2. package/dist/index.d.ts +84 -220
  3. package/dist/index.js +61 -210
  4. package/dist/index.js.map +1 -1
  5. package/docs/feature-guide.md +2 -2
  6. package/docs/wire-protocol.md +1 -1
  7. package/package.json +1 -1
package/docs/feature-guide.md CHANGED
@@ -33,7 +33,7 @@ trying, and whether a change made them better or worse.
33
33
  | “Human feedback should become reusable eval data.” | `FeedbackTrajectory` | Captures approvals, rejections, edits, choices, metrics, and policy blocks. |
34
34
  | “Can this action run, or does it need approval?” | `evaluateActionPolicy` | Generic preflight for side effects, budgets, and required evidence. |
35
35
  | “I need train/dev/test/holdout examples.” | `Dataset` plus feedback trajectory conversion | Stable splits and contamination control. |
36
- | “Which prompt or signature wins?” | `PromptOptimizer`, `OptimizationLoop`, steering optimizers | Runs variants on scenarios and compares scores. |
36
+ | “Which prompt or signature wins?” | `runMultiShotOptimization`, steering optimizers | Runs variants on scenarios and compares scores. |
37
37
  | “Improve a multi-turn agent over real task traces.” | `runMultiShotOptimization` | GEPA-style trajectory optimization with ASI and held-out promotion. |
38
38
  | “Improve prompts, then code if prompts plateau.” | `runPromptEvolution`, composite mutator, code mutator | Bounded evolution with telemetry and lineage. |
39
39
  | “Find why a regression happened.” | bisector, traces, run records | Narrows changes and preserves evidence. |
@@ -156,7 +156,7 @@ Store as `FeedbackTrajectory`, then derive:
156
156
  | Feedback data | `FeedbackTrajectory`, stores, converters | Human/environment labels | Domain adapters live in downstream repos. |
157
157
  | Action policy | `evaluateActionPolicy` | Approval/budget preflight | Blocks or labels actions before `act()`. |
158
158
  | Datasets | `Dataset`, holdout tools, canaries | Train/dev/test/holdout corpora | Keeps optimization honest. |
159
- | Optimization | `PromptOptimizer`, `OptimizationLoop`, steering optimizers | Prompt/signature comparison | Use held-out gates before promotion. |
159
+ | Optimization | `runMultiShotOptimization`, steering optimizers | Prompt/signature comparison | Use held-out gates before promotion. |
160
160
  | Evolution | prompt/code mutators, sandbox pool, telemetry | Autoresearch and mutation loops | Use budgets and lineage; do not run unbounded. |
161
161
  | Telemetry | `TraceStore`, OTLP, file sinks | Audit and replay | Treat traces as evidence, not just logs. |
162
162
  | Reporting | summaries, pareto, cost tracker | Decision support | Useful for PRs, launch gates, research notes. |
package/docs/wire-protocol.md CHANGED
@@ -96,7 +96,7 @@ GET /v1/version
96
96
  ```json
97
97
  {
98
98
  "package": "@tangle-network/agent-eval",
99
- "version": "0.18.0",
99
+ "version": "0.19.0",
100
100
  "wireVersion": "1.0.0",
101
101
  "apiSurface": ["judge", "listRubrics", "version"]
102
102
  }
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@tangle-network/agent-eval",
3
- "version": "0.18.0",
3
+ "version": "0.19.0",
4
4
  "description": "Trace-first evaluation framework for Tangle agents. Core (spans, pipelines, sandbox harness, OTLP export), trust (dataset, red-team, calibration, behavior DSL), builder-of-builders (three-layer eval, resumable sessions, meta-runtime correlation), and frontier (meta-eval correlation study, Process Reward Modeling, bisector).",
5
5
  "homepage": "https://github.com/tangle-network/agent-eval#readme",
6
6
  "repository": {