Seven skills. One loop.
One skill runs the cycle. The others each do one job and hand off to each other. All are open Agent Skills that any coding agent can load.
Run the loop
The entry point. Run it by hand or on a schedule.
- 01kaizen-loopRuns one improvement cycle: measures, picks the single highest-value fix, hands it to the right skill, verifies, records.measure → pick → improve → verify → record
Measure
Scripts produce the numbers. The agent reads them, it doesn't guess.
- 02audit-ci-runsFinds where CI time and money go across many runs: waste, red streaks, flaky jobs, queueing, hygiene. Scores CI health 0–100.ci_runs.py · workflow_hygiene.py
- 03assess-build-cachingScores how well a codebase fits task-level build caching (moon, Turborepo, Nx, Bazel), estimates the speedup and plans the rollout.deps_graph.py · change_locality.py
Fix
One change at a time, verified against every other job.
- 04triage-ci-failureTakes a failing run down to a root cause, then fixes it or files a precise report.failing run → root cause
- 05fix-flaky-testFinds why a test passes and fails on the same commit, and makes it deterministic.flaky → deterministic
- 06speed-up-pipelineMakes the pipeline faster without reducing what it checks: critical path, caching, parallelism, skipping unchanged work.slow step → faster, same coverage
- 07reduce-test-bloatCuts redundant tests with per-test coverage, timings, test impact analysis and mutation testing.collect.py · test_bloat.py
Why continuous, and why now.
Read where we're taking Kaizen CI.