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 instructions/sumulige/sumulige-claude/agents-mdgit clone --depth 1 https://github.com/sumulige/sumulige-claudeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/sumulige/sumulige-claude/agents-md)<a href="https://agentmods.dev/instructions/sumulige/sumulige-claude/agents-md"><img src="https://agentmods.dev/badge/instructions/sumulige/sumulige-claude/agents-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.06196 | $0.06196 |
| Opus 5 | $0.03098 | $0.03098 |
| Sonnet 5 | $0.01239 | $0.01239 |
| Haiku 4.5 | $0.00620 | $0.00620 |
Grade B, and why
sumulige-claude AGENTS.md scanned grade B with 1 finding 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 4d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
位置：`~/.claude/config.json` How it starts
The opening of the file, as written. The whole thing — 643 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Agent 系统详解
从零开始理解多 Agent 协作系统
Technical Co-Founder Working Agreement (Always-on)
- You are my Technical Co-Founder. I am the Product Owner (PO). 你负责执行，我负责决策。
- Keep me in control: stop at key decision points (A-G): tech stack, data model, external services/cost, auth/security/privacy, deploy/domain, UI direction, milestones/timeline.
- Before building, list assumptions as a checklist; I can veto line-by-line. 未确认假设时不要实现。
- If Skill
technical-cofounderis present: use phase gatePHASE=1..5(defaultPHASE=1), only progress one phase per run, and end with numbered confirmation options. - 中文为主，保留关键英文术语；先讲人话，再给技术细节。
Layer 1: 什么是 Agent？ (5 分钟)
最简单的解释
Agent = AI 角色 + 职责 + 专长
就像一个真实团队：
- 不是让一个人做所有事
- 每个人专精某个领域
- 大家协作完成任务
类比：餐厅团队
没有分工的餐厅 有分工的餐厅
──────────────── ────────────────
┌─────────────────┐ ┌─────────────────┐
│ 老板 (全能) │ │ 服务员 │ ← 接待顾客
│ │ │ 厨师 │ ← 做菜
│ ├─ 买菜 │ │ 采购员 │ ← 买食材
│ ├─ 做菜 │ │ 清洁工 │ ← 打扫
│ ├─ 端菜 │ │ │
│ ├─ 收钱 │ │ │
│ └─ 打扫 │ │ │
└─────────────────┘ └─────────────────┘
结果：累死、效率低 结果：专业、高效
Claude Code 中同样适用
传统 Claude Sumulige Claude Agents
─────────── ────────────────────────
┌─────────────────┐ ┌─────────────────────────┐
│ Claude │ │ Conductor (协调) │
│ (通用 AI) │ │ Architect (架构) │
│ │ │ Builder (实现) │
│ ├─ 理解需求 │ │ Reviewer (审查) │
│ ├─ 设计架构 │ │ Librarian (文档) │
│ ├─ 写代码 │ │ │
│ ├─ 审查代码 │ │ │
│ └─ 写文档 │ │ │
└─────────────────┘ └─────────────────────────┘
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.
- 4d ago First seen · 643 lines · 6,196 tokens per session scan B bd41b7bb959a
sumulige-claude AGENTS.md is an instructions file published in the GitHub repository sumulige/sumulige-claude (2 stars, last pushed 6mo ago), licensed MIT. It adds 6,196 tokens to every session, about $0.0310 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
co-evolution CLAUDE.md
Instructions for alanshurafa/co-evolution, covering co-evolution - cross-ai document refinement, default rule, components, co-evolution skill (skills/co-evolution/) and co-evolve bouncer (co-evolve-bouncer.sh).
co-evolution AGENTS.md
Instructions for alanshurafa/co-evolution, covering for agents reading this repo, the two markers, how to interact with this repo, lab features and project context (gsd-managed).
fast-resume AGENTS.md
AGENTS.md instructions for angristan/fast-resume, covering repository instructions, project, behavior, adding or changing an adapter and validation.
crewplane GEMINI.md
Gemini CLI instructions for crewplaneai/crewplane, a project described as: Run reviewable, resumable coding-agent workflows across Claude Code, Codex, Gemini, GitHub Copilot CLI, or any CLI. You define the stages in Markdown; Crewplane runs the workflow, agents do the work, and every handoff stays on disk.
automate CLAUDE.md
Claude Code instructions for kobiton/automate, covering claude.md, getting started, commands, single test file and architecture.
turnbreak AGENTS.md
AGENTS.md instructions for Milkeles/turnbreak, covering agent instructions file: turnbreak, 1. project overview, 2. commands, 3. code style and 4. repository conventions.