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/lvndry/jazz/agents-mdgit clone --depth 1 https://github.com/lvndry/jazzWhat 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 | $0.01547 | $0.01547 |
| Opus 5 | $0.00773 | $0.00773 |
| Sonnet 5 | $0.00309 | $0.00309 |
| Haiku 4.5 | $0.00155 | $0.00155 |
Grade A, and why
jazz 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 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.
How it starts
The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Working on Jazz
Jazz is an agent harness. You are working on the machine that turns a model into something that can actually do a job — unattended, on a real machine, with real consequences. That is hard, interesting work. Treat it that way.
Do not reconstruct Jazz from memory. Do not answer from vibes. The documentation is the source of truth for what Jazz is and how it behaves. The code is the source of truth for how it is implemented. Read both before you speak.
Go read
Start here, in this order, for whatever you are about to do:
- What is Jazz, and what can it do? README.md, docs/
- How does the harness work? docs/internals/
- Why is it built this way? design-decisions.md
- Where does my change go? code-map.md
- What does the interface demand? docs/design/
- What is the security model? SECURITY.md, threat-model.md
- How do I contribute? CONTRIBUTING.md
- Public flags, tools, config? docs/reference/
- Did this harness change help? evals, evals internals
The stack, the commands, the architecture, the Effect patterns, the naming: they live in the repo. Open the files. Match what you find.
Read the code before you answer
A question about Jazz is a request to go look. Trace the path. Read the tests around it. Follow the types. Then answer with what the code actually does, not what a similar project would do.
If you have not opened the relevant files, you are not ready to:
- explain how something works
- propose a design
- estimate scope
- change anything
Guessing is slower than reading. Reading is the job.
Think at the frontier
Jazz exists to find out how much of the gap between a weak model and a strong one is closable by the harness rather than the model. That is the interesting problem. Lean into it.
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 · 152 lines · 1,547 tokens per session scan A f81ab97d5273
jazz AGENTS.md is an instructions file published in the GitHub repository lvndry/jazz (50 stars, last pushed 2d ago), licensed MIT. It adds 1,547 tokens to every session, about $0.0077 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
intelligent-terminal rust.instructions.md
Concise Rust coding conventions for this repository.
InvestSkill GEMINI.md
Instructions for yennanliu/InvestSkill, covering investskill — gemini cli setup & usage guide, installation & setup, quick start, navigate to the investskill directory and start gemini cli (loads gemini.md automatically).
she-love-me CLAUDE.md
Instructions for 863401402/she-love-me, covering claude.md and 唯一工作流.
huly-mcp CLAUDE.md
Instructions for dearlordylord/huly-mcp, covering project instructions, design principle: llm-first api, project harness (copy to new projects), package manager and verification.
wayland-core copilot-instructions.md
Instructions for FerroxLabs/wayland-core, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.
ZipAgent AGENTS.md
Instructions for JiayuXu0/ZipAgent, covering repository guidelines, project structure & module organization, build, test & development commands, coding style & naming conventions and testing guidelines.