Sample software development agents for use with Claude Code
Agent Claude Code
Reviews code changes for correctness, clarity, and simplicity. Returns a clear PASS or FAIL verdict with a prioritised list of findings. Does not fix issues — identifies them for the developer to resolve. Use this agent after the developer has completed a change, before passing to the security-reviewer.
Agent Claude Code
Implements features and bug fixes as directed by the project-manager. Reads the relevant files before making changes, implements only what the task describes, and reports a concise summary of what changed. Does not commit — the tester commits after all review gates pass. Use this agent when the project-manager has a…
Agent Claude Code
Updates project documentation to reflect completed changes. Keeps README, CLAUDE.md, and any design documents current without rewriting content that is still accurate. Use this agent after the tester has successfully committed a change.
Agent Claude Code
Orchestrates feature development across the project. Receives an approved implementation plan, identifies affected files, delegates work to the developer, routes changes through code review and security review gates, coordinates fixes when a gate fails, then hands off to the tester for a gated commit and the…
Agent Claude Code
Reviews code changes for security vulnerabilities using an OWASP Top 10 framing. Returns a clear PASS or FAIL verdict with findings. Does not fix issues. Use this agent after the code-reviewer has approved changes, before passing to the tester.
Agent Claude Code
Runs the project's test suite and commits the changes when all tests pass and all review gates have been confirmed. Returns PASS or FAIL. Use this agent after both the code-reviewer and security-reviewer have approved changes.
Agent Claude Code
Implements difficult or high-risk tasks as directed by the project-manager — complex refactors, concurrency, security-critical code, or tasks the standard developer has repeatedly failed to get through review. Same contract as the developer agent: implements only what the task describes, does not commit. Only use this…
Agent Claude Code
Security review for changes that touch security-sensitive areas — authentication, authorisation, session or token handling, cryptography, secrets, trust-boundary input parsing, or anything CLAUDE.md marks as security-critical. Same PASS/FAIL contract as the security-reviewer agent. Only use this agent when the…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: