Browse AI Coding Agent Skills

Guide debugging with expected behavior, reproduction, root cause analysis, and validation.

Apply test-driven development by writing a failing test before production code.

Require fresh terminal evidence before an agent claims work is complete.

Use AI DevKit memory as a durable knowledge layer for reusable project context.

Exchange information with active Codex, Claude Code, and other AI agents.

Scan running agents, assess progress, and coordinate multi-agent workflows.

Document code entry points with structured analysis and dependency mapping.

Analyze and simplify complex code paths for maintainability.

Review code, skills, and prompts for vulnerabilities and insecure defaults.

Review and improve documentation for clarity, structure, and novice comprehension.

Update CHANGELOG.md from commits since the latest release.

Create safe, verified commits from only the intended changes.

Set up or resume the right workspace before feature work.

Capture and review requirements before design or implementation.

Validate architecture and design trade-offs against requirements.

Create and reconcile implementation plans from requirements and design.

Execute planned implementation and check it against the approved design.

Add and validate test coverage for feature work.

Run a final pre-push review across design, implementation, tests, and risk.

Publish a reviewed feature branch for code review.