mason-context 0.20.1 → 0.21.0

package/CHANGELOG.md

@@ -1,6 +1,17 @@
 # Changelog
 
-## Unreleased
+## 0.21.0 — 2026-10-10
+
+- Continue Claude once at Stop when outstanding decision drift overlaps observed session anchor edits. Require source inspection, preserve human approval boundaries, and keep repeat Stops and Codex advisory. Add local `mason auto config --stop-block off|drift` controls and status reporting. Include uncommitted anchor changes in drift findings for accepted and legacy decisions using Git baselines.
+
+- Capture readable decision anchors when saving proposals, so committing a proposal with its implementing code does not create false drift. Later source edits still require inspection; unreadable scopes retain unknown freshness.
+- Add separate source inspections that resolve covered drift without claiming human acceptance or changing the accepted revision. Bind preparation tokens to source bytes, preserve original audit evidence, and reopen inspections after relevant changes.
+- Batch anchor inventories, index path scopes, and reuse decision reads within each inspection while retaining fresh final validation. Add reproducible validation against pinned Now in Android and WordPress Android repositories, with measured hook costs documented in `docs/hook-performance.md`.
+
+## 0.20.2 — 2026-09-30
+
+- Show decision titles and the inspection command before changed file paths in completion warnings, so long paths cannot hide the reason for review or the next step.
+- Add regression coverage for long Android paths and historical advisory closure after accepting a replacement decision, including automation checks.
 
 ## 0.20.1 — 2026-09-30
 

package/README.md

@@ -86,3 +86,17 @@ Earlier read-only decision-retrieval evaluations scored **9.0/10 with Mason vs 7
 - [Audits, repair verification, and CI evidence](docs/checks.md)
 - [Decisions, tools, and optional architecture maps](docs/reference.md)
 - [Roadmap](ROADMAP.md) · [Changelog](CHANGELOG.md) · [MIT license](LICENSE)
+
+## Development
+
+For development, use Node 20 (20.19+), Node 22 (22.13+), or Node 24+.
+After `npm ci`, run `npm run format` to format TypeScript sources, tests, and tooling configs.
+CI checks formatting with `npm run format:check` and runs `npm run lint`,
+`npm run typecheck`, and the test suite. Run `npm test` locally to check behavior.
+
+ESLint uses the recommended TypeScript rules plus type-aware checks for unhandled
+promises, misused async callbacks, and invalid awaits. Prettier handles formatting.
+Test fixtures and generated outputs are excluded from formatting and linting.
+Production sources also reject explicit `any` and unsafe assignments, property
+access, calls, arguments, and returns. External data is validated before use.
+Test doubles keep the more permissive rules.