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/zhyx12/projtool/claude-mdgit clone --depth 1 https://github.com/zhyx12/projtoolWhat 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 | $0.03126 | $0.03126 |
| Opus 5 | $0.01563 | $0.01563 |
| Sonnet 5 | $0.00625 | $0.00625 |
| Haiku 4.5 | $0.00313 | $0.00313 |
Grade A, and why
projtool 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 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.
This is a copy
92% identical to projtool AGENTS.md — 43 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — for developers working ON projtool
This file is for Claude Code (or any AI agent) when working in this repository — the projtool source tree.
This is NOT the CLAUDE.md that gets shipped to users. That one is at
src/projtool/assets/project_templates/CLAUDE.md.snippet and is data —
it gets injected into user projects during mcp__projtool__retrofit_write_template.
Modifying that file changes user-facing behavior; modifying this file
changes how you work on projtool itself.
What projtool is, in one paragraph
projtool is a Python package + Claude Code MCP server that helps a single
ML researcher run experiments across a local dev machine and remote AutoDL
GPU containers. It manages git worktrees, mutagen sync sessions, AutoDL
instance lifecycle, manifest.yaml bookkeeping, and a paired "docs repo"
where analysis reports live (separate git repository from the code repo).
End users invoke projtool setup once from a shell, then operate through
mcp__projtool__* tool calls inside Claude Code from then on. See
docs/product-spec.md for the full product story.
CRITICAL: you are working ON projtool, not WITH projtool
This is the most common confusion when developing this project. Three rules:
-
The
mcp__projtool__*tools do not exist in this development environment. They are the target you are building. References to them in design docs, slash commands, hooks, and skills undersrc/projtool/assets/are part of the user-facing API surface — they are what we ship, not what we use to build. -
The
ml-research-workflowskill atsrc/projtool/assets/project_skills/ml-research-workflow/does not apply to this project. That skill is for end-user ML research projects that have been retrofitted with projtool. This project is a Python package; you do not have GPUs, training runs, orexperiment_ids. Do not load that skill. -
The
projtool-bootstrapskill atsrc/projtool/assets/user_skills/projtool-bootstrap/also does not apply. It activates in user projects when the user says "set up projtool here". It has no role in developing projtool itself.
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 · 237 lines · 3,126 tokens per session scan A 5d6edc70d588
projtool CLAUDE.md is an instructions file published in the GitHub repository zhyx12/projtool (1 stars, last pushed 19d ago), licensed MIT. It adds 3,126 tokens to every session, about $0.0156 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to projtool AGENTS.md, differing in 43 lines, and is treated as a copy.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.