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/jinplu/teamwork/claude-mdgit clone --depth 1 https://github.com/JinPLu/TeamworkWhat 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 | $0.00621 | $0.00621 |
| Opus 5 | $0.00311 | $0.00311 |
| Sonnet 5 | $0.00124 | $0.00124 |
| Haiku 4.5 | $0.00062 | $0.00062 |
Grade A, and why
Teamwork CLAUDE.md 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
@AGENTS.md
Claude Code compatibility adapter
Claude Code is an explicit compatibility and development target, not part of the default Codex Update or validation path.
./install.sh claude
claude installs the Skills, helper roles, and the managed global policy.
The policy is written to ~/.claude/CLAUDE.md. To print the same block
without writing it, run ./install.sh claude-policy. Invoke Skills with
/name in Claude Code (for example /teamwork-collaborate); Codex uses
$name.
Agent availability and adapter freshness never block ordinary work. It creates no project document schema, Case lifecycle, mandatory Writer workflow, or migration state.
Roles and host modes
AskUserQuestion handles bounded Collaborate batches. Task/Agent dispatches the
seven optional helpers: Researcher, Debugger, Challenger, Planner, Reviewer,
Worker, and Writer. Writer is a Task/Agent helper role, not a Skill. Built-in
Explore handles live local search, with thoroughness quick, medium, or very
thorough; do not name a custom agent Explore, because that identifier
overrides the built-in. Plan mode and the host Plan subagent are not
substitutes for Teamwork Planner. Debugger stays because this host has no
diagnosis mode. Claude still installs Debug, Goal, and Debugger.
The managed block in ~/.claude/CLAUDE.md owns the host mapping: which native
surface is an editable candidate, which signal is acceptance, and which host
paths are machine-local rather than Teamwork persistence — the plan file under
~/.claude/plans/ and auto memory under
~/.claude/projects/<project>/memory/. Accepted reusable results persist under
docs/teamwork/<kind>/
. This host reads CLAUDE.md and not AGENTS.md, so init-project also
writes a small managed @AGENTS.md import and the project block is loaded
through it.
A Cursor install that refreshes this Claude skill root still installs the full
Claude set. When both same-named Teamwork skill copies exist under
~/.cursor/skills/ and ~/.claude/skills/, which copy a dual-host session
reads is not guaranteed—keep both in sync via the installers.
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.
- 3d ago First seen · 56 lines · 621 tokens per session scan A e0f381ee8acf
Teamwork CLAUDE.md is an instructions file published in the GitHub repository JinPLu/Teamwork (11 stars, last pushed 7d ago), licensed MIT. It adds 621 tokens to every session, about $0.0031 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 2026-08-30.
Other instructions, from other repositories
skills AGENTS.md
Instructions for qdrant/skills, covering qdrant skills, project overview, project structure, navigating skills locally and conventions.
agentic-playwright selectors.instructions.md
Instructions for idavidov13/agentic-playwright, covering selector strategy, critical, instructions, phase 1: open and authenticate and phase 2: explore like a user.
nezha AGENTS.md
AGENTS.md instructions for hanshuaikang/nezha, covering nezha — agents.md, 项目概述, 开发命令, 架构设计 and 前端（src/）.
awesome-copilot-id AGENTS.md
Instructions for GulajavaMinistudio/awesome-copilot-id, covering communication, explanation and documentation, markdown formatting, user communication style and workflow & methodology.
llm-ide-rules pytest-integration-tests.instructions.md
Instructions for iloveitaly/llm-ide-rules, covering pytest integration tests and example integration test.
toki AGENTS.md
Instructions for choi138/toki, covering workspace rules, source of truth, toki workflow, required checks and global safety rules.