From mz-dev-pipe
Orchestrates a deep verification pass that checks whether code in scope is correct, clean, well-tested, and functional. This pipeline reports findings — it does NOT auto-fix anything. The user decides what to fix based on the report.
npx claudepluginhub doctormozg/claude-pipelines --plugin mz-dev-pipeThis skill uses the workspace's default tool permissions.
Orchestrates a deep verification pass that checks whether code in scope is correct, clean, well-tested, and functional. This pipeline reports findings — it does NOT auto-fix anything. The user decides what to fix based on the report.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Checks Next.js compilation errors using a running Turbopack dev server after code edits. Fixes actionable issues before reporting complete. Replaces `next build`.
Guides code writing, review, and refactoring with Karpathy-inspired rules to avoid overcomplication, ensure simplicity, surgical changes, and verifiable success criteria.
Share bugs, ideas, or general feedback.
Orchestrates a deep verification pass that checks whether code in scope is correct, clean, well-tested, and functional. This pipeline reports findings — it does NOT auto-fix anything. The user decides what to fix based on the report.
polish or debug.build.audit.$ARGUMENTS — Optional scope and focus. Any combination of:
"src/auth/", "tests/test_payments.py" — which files to verify"check examples work", "verify the API layer" — what to focus on"src/payments/ check all edge cases are tested"If empty, verify the entire project (roam mode with standard exclusions).
See skills/shared/scope-parameter.md for the canonical scope modes (branch, global, working) and their git commands. Document any skill-specific overrides or restrictions below this line.
scope:): path/glob detection from argument, or roam entire project if empty.scope: determines which source files are under verification. Tests always run fully (not filtered) to catch regressions. Coverage and quality analysis focus on code within scope..mz/task/ — working artifacts under .mz/task/<task_name>/| # | Phase | Reference | Loop? |
|---|---|---|---|
| 0 | Setup | inline below | — |
| 1 | Scope Resolution | phases/setup.md | — |
| 2 | Tooling Detection | phases/setup.md | — |
| 3 | Execution | phases/checks.md | — |
| 4 | Coverage & Quality Analysis | phases/checks.md | — |
| 5 | Failure Diagnosis | phases/checks.md | — |
| 6 | Report | phases/checks.md | — |
scope:) into path-like tokens (globs, dirs, files) and focus tokens (free text).<YYYY_MM_DD>_verify_<slug> where <YYYY_MM_DD> is today's date (underscores) and slug is snake_case (max 20 chars); on same-day collision append _v2, _v3..mz/task/<task_name>/, write state.md with Status, Phase, Started.phases/setup.md → Phase 1. Update state to scope_resolved.phases/setup.md → Phase 2. Update state to tooling_detected.phases/checks.md → Phase 3. Update state to checks_executed.phases/checks.md → Phase 4. Update state to analysis_complete.phases/checks.md → Phase 5. Update state to diagnosis_complete..mz/reports/<YYYY_MM_DD>_test_<scope_name>.md (append _v2, _v3 if exists). See phases/checks.md → Phase 6. Update state to completed.Techniques: delegated to phase files — see Phase Overview table above.
| Rationalization | Rebuttal |
|---|---|
| "tests passed last time, skip rerun" | "environment drift silently breaks suites between runs" |
| "type-check is slow, skip" | "the bug you didn't type-check is the one that crashes in staging" |
| "coverage is a vanity metric" | "coverage < 70% means you're shipping blind in the uncovered 30%" |
Output the final report block: per-check pass/fail status, coverage percentages, top quality findings, and the written report path. No silent skips.
After each phase, update .mz/task/<task_name>/state.md with: