ai-agent-config

Subagents

Every subagent in agents/, with the description from its frontmatter. Each one is also rendered to TOML for Codex.

Claude Code reads these from ~/.claude/agents, Codex reads the generated TOML from ~/.codex/agents. The showcase shows each source next to its TOML file.

5 agents, read from agents/ when this site was built.

explorer
Searches a codebase to answer a specific question — "where is X", "how does Y work", "what calls Z". Protects the main agent's context by keeping raw search output isolated and returning a concise synthesis.
implementer
Executes an agreed plan with minimal, reviewable diffs. Preserves existing architecture, does not restructure silently, validates what it changed.
planner
Breaks non-trivial work into a concrete, ordered implementation plan. Surfaces risks, assumptions, and the smallest reasonable path. Does not implement.
reviewer
Reviews a diff or proposed change for correctness, regression risk, maintainability, and architectural fit. Returns findings grouped by severity. Does not implement.
security-auditor
Audits code or diffs for security issues — injection, authn/authz flaws, secret handling, input validation at boundaries, dependency risks. Defensive focus; does not write exploits.

Edit this page on GitHub