Garry Tan 336dbaa50d fix: detect is_error from claude -p result line (ConnectionRefused was PASS) ...

claude -p can return subtype="success" with is_error=true when the API is
unreachable. Previously we only checked subtype, so API failures silently
passed. Now check is_error first and report as 'error_api'.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-14 12:35:43 -05:00

..

eval-store.test.ts

feat: eval persistence with auto-compare against previous run

2026-03-14 03:49:47 -05:00

eval-store.ts

feat: add E2E observability — heartbeat, progress.log, NDJSON persistence, savePartial()

2026-03-14 11:04:16 -05:00

llm-judge.ts

feat: 3-tier eval suite with planted-bug outcome testing (EVALS=1)

2026-03-14 01:17:36 -05:00

observability.test.ts

feat: eval-watch dashboard + observability unit tests (15 tests, 11 codepaths)

2026-03-14 11:04:40 -05:00

session-runner.test.ts

feat: stream-json NDJSON parser for real-time E2E progress

2026-03-14 03:49:36 -05:00

session-runner.ts

fix: detect is_error from claude -p result line (ConnectionRefused was PASS)

2026-03-14 12:35:43 -05:00

skill-parser.ts

feat: 3-tier eval suite with planted-bug outcome testing (EVALS=1)

2026-03-14 01:17:36 -05:00