gretchenfitze/anti-slop / skills/anti-slop-audit
anti-slop-audit
VerifiedAudit a TypeScript or React frontend repo (any linter, test runner, or CI) against the ten anti-AI-slop checks (API contract codegen, strict TypeScript, sonarjs and unnecessary-effect lint rules, layer boundaries, custom rules from past bugs, agent rules and hooks, Stryker mutation testing, knip dead code, jscpd duplicates, CI gates, plus react-doctor): measure what each finds here, decide which are worth it, and write a PR-by-PR rollout plan to anti-slop/plan.md and anti-slop/todo.md. Use when the user wants to harden a repo against AI-generated or vibe-coded code, is turning a prototype, MVP, or vibe-coded app into a product, is starting a project and wants guardrails from day one, asks which linters or dead-code and duplicate detectors their project needs, wants to know how much slop a codebase has, or mentions the Evil Martians anti-slop post. Changes nothing but those two files. To carry out a step of an existing plan, use anti-slop-rollout.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 18.1 KB |
| assets/plan-template.md | markdown | 3.6 KB |
| assets/todo-template.md | markdown | 2.6 KB |
| references/moves.md | markdown | 23.7 KB |
| scripts/probe.mjs | script | 43.2 KB |
assets/todo-template.md
View on GitHub# Anti-slop TODO: <project>
The rollout steps of [plan.md](plan.md), one PR each, in order. The anti-slop-rollout skill takes the first unchecked one. The numbers, the reasoning, and the history are in the plan.
New rules start as warnings that only changed files must pass, and CI gates come last. <The plan's lint backlog variant, one sentence. Gradual: "Old code stays as-is; a warning is promoted once it catches a real bug and turned off after a month of noise." Cleanup: "Cleanup steps take each rule to zero, rules get promoted one at a time, and the rollout ends with no rule at `warn`.">
- [ ] **1. <Title>**: <one-sentence goal>.
- Changes: <packages, files, config keys, scripts>
- Commands: <what the step adds, with the project's package manager, e.g. `npm run deadcode`>
- Done when: <a checkable condition, e.g. "knip reports 0 issues; CI job `deadcode` green">
- Depends on: <earlier step, or none>
- [ ] **2. <Title>**: …
- [ ] **<N>. Review the warnings, a month after they land** (gradual backlog): promote what caught a real bug, turn off what only made noise.
- Changes: the lint config. A rule that caught a real bug (a PR or review names it) moves to `error` once its hits in old code are fixed (a small count here, a large one as its own step). A rule that only made noise is removed. The rest stay warnings until the re-audit.
- Commands: none
- Done when: each promoted rule names its catch in the Log, and a planted hit of it fails lint
- Depends on: the steps that added the rules, merged at least a month earlier
- [ ] **<N>. Clear the rest of <preset> and promote it** (cleanup backlog): every rule of the set becomes an error.
- Changes: fix the <N at audit> hits the earlier cleanup steps left, then move every listed rule to `error`, including those already at zero and `cognitive-complexity` (<N at audit> functions to split). <A rule that can't reach zero here: the follow-up step that promotes it.>
- Commands: none
- Done when: lint prints no warnings, and a planted hit of a promoted rule fails it
- Depends on: the preset's earlier cleanup steps
- [ ] **<last>. Trim the agent rules file**: delete every rule a check now enforces, so the file only says what no check can. Then delete `anti-slop/`.
- Changes: AGENTS.md (or CLAUDE.md, and the other agents' rules files); `anti-slop/` removed
- Commands: none
- Done when: no line restates a check that blocks; the PR lists each deleted line next to the check that replaced it and a plant that check fails on; the required checks match the plan's gates, or the PR names the gap; `anti-slop/` is gone
- Depends on: the last step that adds or promotes a check
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |