Lbstrydom/claude-engineering-skills

A bundle of **AI-pair-programming skills** for planning, auditing, testing, and shipping code. Works with Claude Code, VS Code Copilot, Cursor, Windsurf, and any terminal.
Three-model audit loop (Claude plans/codes, GPT audits, Gemini reviews) with adaptive learning, real-user persona testing against live URLs, and an architectural-memory layer that catches duplicate-function drift before code gets written.
Quick Reference
**Slash commands** (Claude Code / Copilot / Cursor):
> **VS Code Copilot users**: skills load through TWO native mechanisms — the
> committed .claude/skills/** directory is auto-discovered by Copilot's
> Agent Skills
> (the cross-agent open standard; also read by Copilot CLI and the cloud
> agent). Since VS Code 1.109 those skills surface directly as / slash
> commands in Copilot Chat. AGENTS.md is picked up automatically
> as always-on project instructions. No configuration needed.
| When you want to… | Command |
|---|---|
| Design a feature (auto-detects backend / frontend / full-stack) | /plan |
| Run the whole feature flow on autopilot | /cycle |
| Iterate on a plan with GPT + Gemini (max 3 rounds) | /audit-plan |
| Audit code against its plan (multi-pass + Gemini final) | /audit-code |
| QA a deployed app as a persona (Playwright + P0–P3) | /persona-test " " |
| Lock a fix's behaviour with a Playwright spec | /ux-lock |
| Pre-push quality gate (test + lint + commit + push) | /ship |
| "Why is this code shaped this way?" | /explain |
| Brainstorm with multiple LLMs (you drive convergence) | /brainstorm (add --with-gemini) |
| Sync AGENTS.md / CLAUDE.md across all AI agents | /ai-context-management audit |
| Bootstrap or add to per-repo security memory (incident log) | /security-strategy bootstrap · /security-strategy add-incident |
**Architectural-memory** (terminal):
| Command | What it does |
|---|---|
| npm run arch:refresh | Incremental: re-extract changed files, copy-forward the rest |
| npm run arch:refresh:full | Full rebuild from scratch |
| npm run arch:render | Generate docs/architecture-map.md (Mermaid + flat tables) |
| npm run arch:drift | Compute drift score for active snapshot |
| npm run arch:duplicates | List top cross-file duplicate clusters (refactor targets) |
| npm run security:refresh | Re-parse docs/security-strategy.md and refresh the security_incidents Supabase index (auto-runs post-push from /ship) |
**Requirements layer** (terminal) — a materialized view of the codebase's de-facto invariants, surfaced to /audit-code as a rubric:
| Command | What it does |
|---|---|
| node scripts/requirements.mjs extract --files [--runs N] | Extract de-facto requirements from the given files (LLM ×N, merged) → .requirements/candidates.json + gaps.json |
| node scripts/requirements.mjs reconcile | Fold candidates + gap assessments + hand-curated .requirements/overrides.json into .requirements/ledger.json |
| node scripts/requirements.mjs index | Print the active requirements index |
| node scripts/requirements.mjs render | Render the ledger as a human-readable map → docs/requirements-map.md (Mermaid pie + grouped tables) |
Once .requirements/ledger.json exists, /audit-code automatically injects in-scope invariants as a rubric. See .requirements/README.md.
**Workflow**:
---
Architecture
See docs/architecture-map.md — generated index of every symbol in this repo, grouped by domain, with Mermaid diagrams. Regenerated on /ship or via npm run arch:render.
Quick Start
**Add the bundle to a repo** — one command, and the same one to update it later:
That installs the skills into /.claude/skills/, the runners into
/scripts/.claude-skills/, maintains the managed .gitignore block, and
prompts for API keys. Nothing is written to your home directory. Add --dry-run
to see what would change.
**Use npx here even if your project is on pnpm, yarn or bun.** npm ships with
Node, so npx is present regardless of which manager your project uses, and it
is the only spelling of this command that works: pnpm dlx github:… fails with
ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED (verified 2026-08-15). Your project's own
package manager is detected separately — see
Package managers below.
> **Know what that command does.** It fetches this repo and runs its prepare