Borrowing it
Nothing to install: this file belongs to doug/jjj. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/doug/jjj/main/AGENTS.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/agents-md)<a href="https://agentmods.dev/instructions/doug/jjj/agents-md"><img src="https://agentmods.dev/badge/instructions/doug/jjj/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/doug/jjj/agents-md"><img src="https://agentmods.dev/badge/instructions/doug/jjj/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00893 | $0.00893 |
| Opus 5.5 | $0.00357 | $0.00357 |
| Sonnet 5.5 | $0.00179 | $0.00179 |
| Haiku 4.5 | $0.00089 | $0.00089 |
Grade A, and why
jjj AGENTS.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 6d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent instructions
This repository is jjj, a project tracker built on Jujutsu. It is also
managed with jjj — problems, solutions, and critiques live in the jjj
bookmark alongside the code.
Before you write anything
Set an identity. Without it every agent shares the repository owner's identity, so work claims overwrite each other and per-agent queues do not work:
export JJJ_USER=<your agent name>
jjj whoami --json # confirm it took effect
Read this first
skills/jjj/SKILL.md — how to use and collaborate through jjj: the machine
interface, the single-agent loop, and the multi-agent patterns (work claiming,
rival conjectures, adversarial review, collision avoidance, pods). It is checked
against the real CLI by tests/skill_test.rs, so it does not drift.
For working on jjj's own code, CLAUDE.md covers the architecture, storage
model, and build commands.
House rules
- Use ids, not titles, in anything scripted. Fuzzy title matching resolves to
whatever uniquely matches today. Read ids from
--json. - Never
--forcepast a critique.jjj solution approve --forcebypasses the one mechanism that makes review real. - Automation rules are machine-local (
.jj/jjj-meta/automation.toml) and are deliberately never synced. Do not move them intoconfig.toml; that file travels through the shared bookmark and a rule there would execute on every collaborator's machine. jjj primeis the first call of a turn. It gives you what to work on, what is already proposed, what has already been measured, and what is blocked on a person — in one call with output that does not grow with the corpus. Re-deriving that with a dozen listings is the single largest waste in a swarm run.- Guard every write you based on a read. Pass
--if-lamport <N>with the clock fromshow --json(jq '.lamport // 0'— a clock of 0 is omitted). A refused write exits 13; treat that as "re-read and retry", and only exit 1 as "the command was wrong". - Record evidence as a finding, not in prose.
jjj finding newwith--methodis what stops the next agent re-measuring what you already measured, and--measured-atkeeps two findings a week apart from looking simultaneous. jjj escalateis only for what a human must do — a dead credential, a destructive step, a blocker the playbook does not cover. Not retries, not routine progress. A channel carrying those is a channel nobody reads.jjj doctoris the first thing to run when something looks wrong, and the right thing to attach to a bug report.
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.
- 6d ago Changed · +33 lines · +429 tokens per session 925b5fa886a4
- 2mo ago First seen · 51 lines · 464 tokens per session scan A abe20002c3c7
jjj AGENTS.md is an instructions file published in the GitHub repository doug/jjj (13 stars, last pushed 7d ago), licensed Apache-2.0. It adds 893 tokens to every session, about $0.0036 per session on Opus 5.5. 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
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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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 copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.