--- description: Run every pre-merge gate in parallel…
Community ralph loop for testing, sourced from github. Verified exit condition, evaluator-gated.
--- description: Run every pre-merge gate in parallel and report a punch-list allowed-tools: Bash(pnpm: ), Bash(git: ), Bash(jq: ), Bash(grep: ) --- Verify the current branch is mergeable. Run every gate in parallel and surface a single consolidated report. ## Gates (run all in parallel) 1. Typecheck — pnpm typecheck . Must be clean. 2. Tests + coverage — pnpm test . Coverage must be 100% on every metric (statements / branches / functions / lines). Anything below is a fail. 3. Lint — pnpm lint . Warnings are fine; errors fail the gate. 4. Build — pnpm build . Must complete without TS errors. 5. Charter-covered diff — git diff master --stat -- libs/act/src/builders/ libs/act/src/act.ts libs/act/src/types/ports.ts libs/act/src/types/index.ts libs/act/src/ports.ts . If any file in that list changed, explicitly note "charter surface modified — categorize as additive/breaking before merging." 6. Doc-staleness audit — run the doc-grep recipe from CLAUDE.md "Pre-handoff workflow → Doc audit": bash # Identify renamed/removed identifiers in this PR git diff master --diff-filter=D --name-only -- libs/act/src/ libs/act- /src/ # For each r
claude-code
More testing loops
Ship the release
Run tests and update changelog until both pass without altering the public API.
/goal ship the release --success "tests pass and changelog updated" --constraints "do not touch the public API" --mode ordered
testingmedium risk
/ralph-loop "fix all tests" --max-iterations 5 (iterates until…
Community ralph loop for testing, sourced from github. Verified exit condition, evaluator-gated.
/ralph-loop "fix all tests" --max-iterations 5 (iterates until passing
testingmedium risk
Trim what loads before first paint
Reduce the data downloaded before the first screen appears, with tests and screenshots guarding behavior and appearance.
Reduce the data [web app] downloads before its first screen appears. First record passing tests, mobile and desktop screenshots, and compressed transferred bytes—the data actually downloaded. Use the build report only to suggest candidates. Defer, compress, or remove one item, then rebuild and rerun every check. Keep it only if tests pass, screenshots are pixel-identical, and bytes decrease; otherwise revert. Stop when no safe candidate remains, progress stalls, or approval is needed. Return measurements, changes, and untested states.
testingmedium risk