Skip to content

Commit 6d496e6

Browse files
committed
chore(release): 0.19.1
1 parent 7e875ae commit 6d496e6

2 files changed

Lines changed: 13 additions & 1 deletion

File tree

CHANGELOG.md

Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,17 @@
11
# Changelog
22

3+
## 0.19.1 — release confidence gate
4+
5+
### Added
6+
7+
- `evaluateReleaseConfidence`, a conservative release scorecard over corpus
8+
coverage, search/holdout run evidence, ASI diagnostics, overfit checks, and
9+
cost/latency budgets.
10+
- `assertReleaseConfidence`, a throwing variant for CI/release scripts.
11+
- `releaseTraceEvidenceFromMultiShotTrials`, a helper that projects
12+
`MultiShotTrialResult` rows into release trace evidence so single-shot and
13+
variable multi-shot apps use the same release gate.
14+
315
## 0.19.0 — legacy optimizer removal
416

517
### Removed

package.json

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
{
22
"name": "@tangle-network/agent-eval",
3-
"version": "0.19.0",
3+
"version": "0.19.1",
44
"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).",
55
"homepage": "https://github.com/tangle-network/agent-eval#readme",
66
"repository": {

0 commit comments

Comments
 (0)