- Optimizes regex parsing for test failures by caching the output string.
- Adds explicit security guardrails to the PR review comment when sensitive files (Nostr, crypto, storage) are modified.
- Cleans up temporary debug files.
- Addresses previous code review feedback regarding completeness and safety.
Co-authored-by: PR0M3TH3AN <[email protected]>
- Optimize regex execution loop by caching the test output string.
- Ensure `debug_regex.js` is removed.
- Verified local build and lint pass.
Co-authored-by: PR0M3TH3AN <[email protected]>
- Parse `npm run test:unit` output to identify and list failing test files in the review comment.
- Add a suggestion to run `npm run lint:css -- --fix` if stylelint errors are detected.
- Maintain existing guardrails and micro-fix logic.
Co-authored-by: PR0M3TH3AN <[email protected]>
- Refactored `scripts/automated-pr-review.mjs` to separate review logic into `reviewPR` function.
- Added `--all` flag to support iterating over all open PRs using `gh` CLI.
- Updated commit message for micro-fixes to `fix(ai): formatting (agent) suggested`.
- Improved error handling and context management (e.g., passing `baseRef`).
Co-authored-by: PR0M3TH3AN <[email protected]>
- Update `scripts/automated-pr-review.mjs` to include missing sensitive paths (`js/auth/`, `js/magnetUtils.js`, `js/nostr/adapters/nip07Adapter.js`) and improve commit messaging for automated fixes.
- Add release channel guidance checks for `unstable` branch in `automated-pr-review.mjs`.
- Create `scripts/agent/review-open-prs.mjs` to robustly batch process open PRs using the updated review script.
- Remove redundant scripts `scripts/agent/automated-pr-review.mjs` and `scripts/agent/review-pr.mjs`.
Co-authored-by: PR0M3TH3AN <[email protected]>
Currently, if the automated PR review script encounters failures (lint/test errors), it exits with code 1, but the failure details are only visible in the posted PR comment. In CI environments where the comment might not be immediately accessible or if posting fails, the logs are unhelpful.
This change unconditionally prints the generated review report (including failure details) to `stdout` at the end of the script execution. This ensures that failure reasons are always visible in the CI logs.
Co-authored-by: PR0M3TH3AN <[email protected]>
- Wrap `main` execution in try-catch block to ensure process exits with code 1 on failure.
- Update `runCommand` to throw meaningful errors on spawn failure (e.g. command not found).
- Add explicit check for `gh` CLI availability in CI environment.
- Improve git command error logging in `getChangedFiles` to capture stderr (e.g. "no merge base").
Co-authored-by: PR0M3TH3AN <[email protected]>
- Updated `scripts/automated-pr-review.mjs` to align with specific prompt requirements for formatting fixes and guardrails.
- Added `js/utils/storage.js` to sensitive paths for security review flagging.
- Created `scripts/agent/review-open-prs.mjs` to iterate over open PRs and run the review script using `gh` CLI.
- Add sensitive path checks for storage/crypto services
- Add checks for modifications to critical config when targeting main
- Implement fallback for micro-fixes: create follow-up PR if push fails
- Add Audit Log to review comment
- Add `scripts/automated-pr-review.mjs` to run format, lint, tests, and security checks.
- Add `.github/workflows/agent-pr-review.yml` to trigger the review on PRs.
- Script flags sensitive file changes (crypto/storage) and posts comments.
- Script attempts to commit formatting micro-fixes if permissions allow.
