ai-lab-agent-skills

Use when reviewing code for bugs with finder, challenger, and referee roles, especially when independent review is needed to reduce false positives.

Optimize an agent skill through bounded repeated trials, binary evaluations, targeted prompt mutations, and auditable keep/discard decisions. Use when asked to optimize, benchmark, evaluate, or systematically improve an existing skill. Produces an improved candidate skill, run artifacts, a results log, and a mutation changelog without overwriting the original.

Review uncommitted git changes for bugs/regressions via Codex MCP and present a structured report. Use when asked to review local changes or find bugs in current work. Requires Codex MCP to be configured.

Organize raw project documentation into a focused, cross-linked docs/ structure with an adaptive INDEX.md and a recommended AGENTS.md entry point. Preserve the source's natural document types; create stories only when they already exist and incremental slices only when the user wants them. Use when starting from raw documentation or restructuring scattered project docs.

Record one web application feature flow as a compact GIF using an existing Playwright installation. Use when the user asks for a visual demo, changelog artifact, documentation asset, or pull-request evidence. Creates local capture artifacts first; installation, publication, commits, uploads, and pull-request edits always require separate approval.

Coordinate specialized agent roles across complex tasks using dependency-aware planning, explicit ownership, structured handoffs, safe integration, and evidence-based quality gates. Use when work has multiple independent streams or genuinely benefits from isolated expertise; fall back transparently when the runtime cannot provide separate workers.

Draft and, after approval, capture an observed reusable lesson in an existing skill's gotchas.md. Use after a verified bug, architecture error, process failure, or misunderstanding when the lesson can be generalized without exposing project or user information.

Run a portable, evidence-based review of the changes intended for merge. Use before merging a branch to inspect correctness, security, maintainability, tests, documentation, and review coverage without modifying the changes.

Interactively review and consolidate existing gotchas.md files within an explicitly selected skills root. Use when lessons have accumulated, overlap, conflict, contain stale or sensitive context, or need clearer scope and conditions.

Interactively resolve open questions and documentation gaps in project docs. Scans all docs for unknowns, proposes resolutions from existing sources, and interviews the user for the rest. Writes answers back into the correct doc files. Use when: onboarding to a project with incomplete docs, preparing docs for a structured specification workflow, or stress-testing a design by walking through every open question.

Create well-structured agent skills following cross-platform best practices. Use when building a new skill, improving an existing one, or creating automation for a recurring workflow.

Audit rules and skills for semantic contradictions, redundancy, and staleness, then interactively resolve with the user. Use periodically when agent performance degrades or after adding many rules/skills.

Create and run Maestro UI test flows for the app in this repo. Use when writing Maestro YAML flows or verifying app behavior on a device/emulator.

Design and implement REST APIs with proper routing, validation, error handling, and documentation. Use when: building backend services, microservices, or adding API endpoints to existing applications.

Implement authentication and authorization from scratch. Covers signup, login, sessions, JWT, role-based access, and protected routes. Use when: adding auth to a new or existing app.

Systematic approach to finding and fixing bugs in any codebase. Use when: debugging errors, investigating unexpected behavior, fixing failing tests, or resolving production issues.

Automated code review for security, quality, and performance. Catches bugs, vulnerabilities, and anti-patterns before they ship. Use when: reviewing PRs, auditing code before release, or checking your own work.

Design database schemas, write migrations, and model relationships. Use when: starting a new project that needs a database, adding tables, designing relationships, or optimizing queries.

Auto-generate documentation from code. Creates READMEs, API docs, and architecture overviews. Use when: a project lacks documentation, onboarding new team members, or preparing code for open source.

Build modern frontend applications with React, Next.js, and Tailwind CSS. Use when: creating web apps, UI components, landing pages, dashboards, or any browser-based interface.

Complete full-stack development with frontend, backend, database, and deployment. Use when: building apps that need both a UI and a server, APIs with a database, or any project that spans the entire stack.

Create polished, conversion-optimized landing pages quickly. Use when: building marketing pages, product launches, portfolio sites, or any single-page website that needs to look professional fast.

Craft and optimize prompts for better AI output. Improve clarity, structure, and effectiveness of prompts for Claude Code, Codex, and other AI tools. Use when: getting poor results from AI, building reusable prompt templates, or creating skills and agents.

Clean up and improve existing code safely. Restructure without changing behavior. Use when: code is messy, duplicated, hard to read, or needs modernization.

Generate new reusable skills from git history, coding patterns, and team practices. The meta-skill: teaches the AI to create new skills. Use when: you have recurring patterns worth capturing, want to codify team knowledge, or need to create custom workflows for your project.

Full test-driven development cycle: Red-Green-Refactor. Write tests first, then implement, then clean up. Use when: building new features with high confidence, or enforcing disciplined development practices.

Write comprehensive tests for existing code. Covers unit, integration, and E2E tests. Use when: adding tests to untested code, improving coverage, or ensuring a feature works before shipping.