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/doug/jjj/claude-mdgit clone --depth 1 https://github.com/doug/jjjWrote 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/doug/jjj/claude-md)<a href="https://agentmods.dev/instructions/doug/jjj/claude-md"><img src="https://agentmods.dev/badge/instructions/doug/jjj/claude-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 | $0.02133 | $0.02133 |
| Opus 5 | $0.01066 | $0.01066 |
| Sonnet 5 | $0.00427 | $0.00427 |
| Haiku 4.5 | $0.00213 | $0.00213 |
Grade A, and why
jjj 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 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.
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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
jjj (Jujutsu Juggler) is a distributed project management and code review system built for Jujutsu (jj). It implements Popperian epistemology (Problems → Solutions → Critiques) as a Kanban-style workflow with no central server, no database, and offline-first operation.
Key insight: jj's stable Change IDs persist across rebases/squashes, allowing metadata to survive history rewrites that would orphan Git commit-based metadata.
Build Commands
Rust CLI
cargo build # Debug build
cargo build --release # Release build
cargo test # Run all tests
cargo test <test_name> # Run specific test
cargo run -- <command> # Execute jjj command
cargo fmt # Format code
cargo clippy # Lint
VS Code Extension
cd vscode
npm install
npm run compile # TypeScript compilation
npm run watch # Watch mode
npm run test # Run mocha tests
npm run lint # ESLint
npm run package # Create .vsix
npm run install-ext # Install locally
Documentation
cd docs-site && npm run dev # Serve locally
cd docs-site && npm run build # Build docs
Architecture
Core Model
- Problems: Things to solve (can form DAG via parent_id). Support tags for categorization.
- Solutions: Conjectures attached to jj Change IDs (not commit hashes). Support tags for categorization.
- Critiques: Error-elimination feedback that blocks solution approval
- Milestones: Time-based goals grouping problems
Storage: Shadow Graph
All metadata lives in an orphaned jjj bookmark, never touching the working copy. Sync via jj git push -b jjj.
problems/{uuid}.md
solutions/{uuid}.md
critiques/{uuid}.md
milestones/{uuid}.md
config.toml
events.jsonl
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 · 171 lines · 2,133 tokens per session scan A 5d28795ad259
jjj CLAUDE.md is an instructions file published in the GitHub repository doug/jjj (11 stars, last pushed 4d ago), licensed Apache-2.0. It adds 2,133 tokens to every session, about $0.0107 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
kaneo AGENTS.md
AGENTS.md instructions for usekaneo/kaneo, covering kaneo agent guide, principles, architecture, boundaries that must hold and follow a change through.
kaneo CLAUDE.md
Claude Code instructions for usekaneo/kaneo, a project described as: 🎯 All you need. Nothing you don't. Open source project management that works for you, not against you.
veritas-kanban AGENTS.md
AGENTS.md instructions for BradGroux/veritas-kanban, covering agents.md — canonical agent instructions for veritas kanban, runtime requirements, repository layout, essential commands and install.
veritas-kanban CLAUDE.md
Claude Code instructions for BradGroux/veritas-kanban, covering claude.md — claude-specific supplement for veritas kanban, what changed in v2.1, lessons learned (claude-specific), security and architecture.
kanban CLAUDE.md
Claude Code instructions for kanban-rs/kanban, covering claude.md, project overview, architecture philosophy, solid principles applied and workspace structure.
FlowBoard AGENTS.md
Instructions for rasimme/FlowBoard, covering flowboard — agent entry point, start here and where things live.