Agent
Primary implementer for any coding task — implementation, refactoring, debugging, code review.
2.0k 19d ago A 22 tokens
original MIT
critic
02Agent
Ruthless post-implementation verifier — rejects work that doesn't meet spec. Read-only except running checks.
2.0k 19d ago A 23 tokens
original MIT
debugger
03Agent
Systematic root-cause diagnosis — reproduce, hypothesize, test, fix, verify.
2.0k 19d ago A 21 tokens
original MIT
documenter
04Agent
Writes and updates docs — READMEs, inline comments, usage examples — matching the project's existing style.
2.0k 19d ago A 24 tokens
original MIT
Agent
Catch-all agent for open-ended, multi-step tasks — research, content authoring, and text transformation (e.g. remastering/rewriting a section per a prompt or spec). Use when no more specific agent fits.
2.0k 19d ago A 49 tokens
original MIT
librarian
06Agent
External docs and library best-practices lookup — official references, real-world examples, GitHub repo discovery.
2.0k 19d ago A 26 tokens
original MIT
oracle
07Agent
Read-only architecture advisor — deep analysis, hard debugging, security and performance consulting.
2.0k 19d ago A 18 tokens
original MIT
planner
08Agent
Read-only; researches the codebase and produces a numbered, verifiable step plan before implementation.
2.0k 19d ago A 21 tokens
original MIT
qa-tester
09Agent
Writes tests, builds test suites, and discovers edge cases across unit, integration, and E2E levels.
2.0k 19d ago A 26 tokens
original MIT
red-team
10Agent
Adversarial security reviewer — find vulnerabilities, injection risks, exposed secrets, and failure modes. Read-only probing.
2.0k 19d ago A 26 tokens
original MIT
scout
11Agent
Fast read-only codebase recon — find files, patterns, functions, and entry points.
2.0k 19d ago A 21 tokens
original MIT