Read-only audit of `.claude/` structural overhead. Surfaces what is loaded into every Claude Code session and how much of the context window it consumes, then recommends the highest-leverage trims.

argument-hint: "[official|beta] [--skip-tests] [--dry-run]" allowed-tools: - Bash - Read - Write - Edit - Grep - Glob - Agent - AskUserQuestion - WebSearch source: gstack/claudekit-engineer ---

Generate [llms.txt](https://llmstxt.org/) files — LLM-friendly markdown indexes of project documentation.

CRITICAL: MUST run for EVERY message. Detects agent, complexity, AND model automatically. Always runs FIRST.

REST/GraphQL API design: resource naming, HTTP methods, status codes, pagination, versioning, error formats. Use for 'design API', 'endpoint patterns'.

Measures harness performance against a small set of ground-truth tasks. Provides the empirical signal that the dead-weight audit (`docs/dead-weight-audit.md`) consumes to make load-bearing decisions about each harness component.

Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence. Use when asked to "open in browser", "test the site", "take a screenshot", or "dogfood this".

Build error triage: detect language, load fix patterns, auto-retry. Use for 'build failed', 'fix compilation', 'type error'. Chains into meow:verify.

Safety guardrails for destructive commands. Warns before rm -rf, DROP TABLE, force-push, git reset --hard, kubectl delete, and similar destructive operations. User can override each warning. Use when touching prod, debugging live systems, or working in a shared environment. Use when asked to "be careful", "safety mode", "prod mode", or "careful mode".

Use when copy-catting, replicating, or adapting features from external systems, repos, apps, or ideas into the current project. Triggers on 'copy this from', 'replicate', 'clone this feature', 'how does X do Y', 'chom', 'like how X does it', 'port from', 'build like', '1:1 copy'.

Use when reviewing code for quality, enforcing KISS/DRY/YAGNI principles, or cleaning up over-engineered abstractions. Pragmatic standards: concise, direct, no unnecessary comments.

Use when a user wants to understand, extract requirements from, or analyze a Confluence page or specification before building. Also triggers when a spec needs to be read for planning or ticket creation. Triggers on 'read the spec', 'analyze this confluence page', 'what does the PRD say', 'parse the requirements'.

ALWAYS activate this skill before implementing EVERY feature, plan, or fix.

Chief Security Officer mode. Infrastructure-first security audit: secrets archaeology, dependency supply chain, CI/CD pipeline security, LLM/AI security, skill supply chain scanning, plus OWASP Top 10, STRIDE threat modeling, and active verification. Two modes: daily (zero-noise, 8/10 confidence gate) and comprehensive (monthly deep scan, 2/10 bar). Trend tracking across audit runs. Use when: "security audit", "threat model", "pentest review", "OWASP", "CSO review".

Operational decision frameworks: triage, escalation, case management, billing ops. Use for 'how to handle X cases', 'build triage', 'escalation protocol'.

Use when a project has no docs/ directory or needs initial documentation generated from codebase analysis. Triggers on "init docs", "create documentation", "generate docs", "docs init", or when docs/ is empty. Do NOT use for updating existing docs (use meow:document-release) or for reference guides (use meow:documentation).

Post-ship documentation update. Reads all project docs, cross-references the diff, updates README/ARCHITECTURE/CONTRIBUTING/CLAUDE.md to match what shipped, polishes CHANGELOG voice, cleans up TODOS, and optionally bumps VERSION. Use when asked to "update the docs", "sync documentation", or "post-ship docs". Proactively suggest after a PR is merged or code is shipped.

Use when updating documentation after code changes, generating changelogs, or syncing API docs. Activates during Phase 6 (Reflect) or on explicit doc requests.

Structured second-pass reasoning after code review or plan evaluation. Offers named reasoning methods (pre-mortem, inversion, red team, Socratic, etc.) to re-examine findings through a specific lens. Use after /meow:review verdict or any time deeper analysis is needed on an existing output.