Claude Code Agents Collection
A collection of specialized Claude Code agent definitions for code review, implementation verification, testing and quality assessment, each packaged as a single markdown configuration file.
Claude Code Agents Collection is a set of specialized agent definitions for Claude Code, Anthropic's coding assistant, each extending it with a focused behavior for a common development task. Agents in the collection include one that verifies implemented code actually matches project specifications, one that checks code changes follow the guidelines written in a project's CLAUDE.md file, one that reviews code for over-engineering and unnecessary complexity, one that gives a realistic assessment of how complete a project actually is against what was claimed, one that verifies claimed task completions are functionally working rather than stubbed, and one for comprehensive UI testing.
Each agent is defined in its own markdown file and is invoked within Claude Code to bring that specific review behavior into a session, such as verifying an authentication system against its stated security requirements or confirming a database integration actually works end to end.
It is licensed under MIT.