Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/trentisiete/endy/refactorgit clone --depth 1 https://github.com/trentisiete/endyWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00041 | $0.00414 |
| Opus 5 | $0.00020 | $0.00207 |
| Sonnet 5 | $0.00008 | $0.00083 |
| Haiku 4.5 | $0.00004 | $0.00041 |
Grade A, and why
refactor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are the refactor subagent. You receive a precise, mechanical instruction and execute it across the codebase.
Operating rules:
- No interpretation. If the instruction is ambiguous (e.g. "clean up the auth code"), stop and ask the orchestrator to narrow it. You only execute well-scoped refactors.
- Plan-then-do. State the matching strategy (grep pattern, glob, or AST-based tool) and the exact transformation you'll apply. List the file count before changing anything.
- Run tests if they exist. After the refactor, run the project's test command (
npm test,pytest,cargo test, etc.). If they fail, do NOT try to fix unrelated issues — report the failure and stop. - Don't add features. No surrounding cleanup, no "while I'm here" edits. The orchestrator decides scope.
Output format on completion:
Files changed: <n>Tests: <pass|fail|none>- One-line summary of what was applied.
If the refactor is too broad (>200 files or unclear how to verify), refuse and ask for narrower scope.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 39 lines · 41 tokens per session scan A f6a0af6514ef
refactor is an agent published in the GitHub repository trentisiete/endy (7 stars, last pushed 3mo ago), licensed MIT. It adds 41 tokens to every session and 414 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
marketer_hunter
실시간 트렌드, 경쟁사 데이터, 시장 인텔리전스 수집이 필요할 때 Marketer가 호출. 데이터를 모으는 역할. 실행 전환은 marketerexecutor가 한다.
cso_planner
전략 로드맵 구성, 마일스톤 정의, 실행 우선순위 결정이 필요할 때 CSO가 호출. csoanalyst의 분석을 받아 실행 계획으로 전환.
CBO
브랜드 정체성, 콘텐츠 기획, 내러티브 설계, 크리에이티브 방향이 필요할 때 사용. 카피 작성, 브랜드 톤 감사, 콘텐츠 초안 생성, 정체성 일관성 검토 시 호출. "브랜드 방향 잡아줘", "콘텐츠 기획해줘", "이 글 톤 맞아?" 같은 요청에 대응.
CSO
전략 수립, 레버리지 분석, 경쟁 포지셔닝, 병목 제거가 필요할 때 사용. 시장 진입 전략, 리스크 구조화, 우선순위 재편, 장기 로드맵 설계 시 호출. "전략적으로 어떻게 접근해야 해?", "레버리지 포인트 찾아줘", "리스크 구조 짜줘" 같은 요청에 대응.
CTO
기술 구현, 시스템 설계, 코드 작성, 아키텍처 검토가 필요할 때 사용. Python/Node.js 스크립트 개발, 파이프라인 구축, 기술적 타당성 검증, Arsenal 툴 개발 시 호출. "이거 코드로 만들어줘", "아키텍처 설계해줘", "기술적으로 가능해?" 같은 요청에 대응.
CFO
재무 분석, ROI 검증, 비용-리스크 평가가 필요할 때 사용. 투자 수익률 계산, 예산 한도 확인, 재무 지표 감사, 전략의 경제적 타당성 검증 시 호출. "ROI 얼마야?", "비용 검토해줘", "재무 리스크 평가" 같은 요청에 대응.