Loop/goalTestingmedium riskintermediatesafety C · 65GitHub

Ship the release

Run tests and update changelog until both pass without altering the public API.

Prompt
→ Claude Code
/goal ship the release --success "tests pass and changelog updated" --constraints "do not touch the public API" --mode ordered
claude-code

Use this when

You are cutting a release and the definition of done is mechanical — tests green and changelog current — but you do not want the agent fixing things by changing the public API along the way.

Done when

✓ Exit requires both success conditions to hold at once. Still check the diff for API-surface changes — the constraint is an instruction, not a type-checker. Keep the actual publish/tag/deploy step for a human.

Why it works

/goal encodes both the success condition (tests pass and changelog updated) and a constraint (do not touch the public API), so the loop can iterate on failures without widening its own scope. --mode ordered makes it get tests green before touching the changelog.

Source: willytop8 ↗graded C · 65/100 — how grades work (Safety Score v2) →

/goalTestingmedium risknew

Fix the flaky checkout test

Run checkout tests repeatedly until 20 consecutive passes, logging failure modes and hypotheses after each iteration.

prompt
→ Claude Code
/goal Reproduce and fix the flaky checkout test so it passes 20 consecutive runs, verified by npm test -- checkout, while preserving checkout API behavior and existing correctness tests. Use the checkout test files, checkout service code, fixtures, and related logs. Between iterations, record the observed failure mode, hypothesis, change, and rerun result. If reproduction stays unavailable or valid paths are exhausted, stop with reproduction attempts, logs, most likely absent condition, blocker, and next input needed
/goalTestingmedium risknew

Fix flaky tests, get CI green

Run an agent to identify and fix flaky tests until your CI pipeline passes completely.

prompt
→ Claude Code
/goal fix all flaky tests and make CI green in a new chat. Pair it with a custom mode to follow a playbook, or /loop for recurring check-ins
/goalTestingmedium risk

Reach a coverage target

Add focused tests for the least-covered files and re-measure each turn until coverage hits 80 percent or the turn cap.

prompt
→ Claude Code
/goal test coverage is at least 80% with all tests passing — add focused tests for the least-covered files, re-run coverage each turn, stop at the threshold or after 12 turns