skills
View on GitHubReusable AI agent skills for software factories: explore ideas, write specs, implement, review, and run autonomous research. Works with Claude Code, Codex, and other skill-compatible agents.
A reusable skill pack for AI coding agents that supports feature discovery, spec writing, implementation, research, and code and visual review. Install skills into compatible harnesses such as Claude Code, Codex, opencode, and Cursor.
Use Cases
Explore unknowns and clarify feature requirementsWrite implementation-ready specificationsBuild features through independently verifiable slicesRun autonomous coding loopsDelegate implementation or review to another agentExperiment systematically to improve a measured outcomeReview code and audit implementation decisionsCompare screenshots and critique visual changesCreate and evaluate agent skills
Built With
- Language
- TypeScript
- Frameworks
- Claude Code · Codex · opencode · Cursor
Tags
agent-skills · software-engineering · spec-driven-development · autonomous-agents · code-review · visual-review · autonomous-research · testing · implementation · Claude Code · Codex