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/ttktjmt/mjswan/agents-mdgit clone --depth 1 https://github.com/ttktjmt/mjswanWrote 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/ttktjmt/mjswan/agents-md)<a href="https://agentmods.dev/instructions/ttktjmt/mjswan/agents-md"><img src="https://agentmods.dev/badge/instructions/ttktjmt/mjswan/agents-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 | $0.00464 | $0.00464 |
| Opus 5 | $0.00232 | $0.00232 |
| Sonnet 5 | $0.00093 | $0.00093 |
| Haiku 4.5 | $0.00046 | $0.00046 |
Grade A, and why
mjswan 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 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.
What it actually says
AGENTS.md
Project
mjswan packages browser-based MuJoCo simulations with real-time policy control into an interactive static web app. It has two sides:
- Python (src/mjswan/) —
Builder/Project/SceneAPI that bundles models, policies, and UI config into a static site. - Frontend template (src/mjswan/template/) — TypeScript + three.js + mujoco-wasm client that the Python build step bundles.
See CONTEXT.md for a full codebase map, object model, module descriptions, and tooling reference.
Philosophy
- Write clean, readable, maintainable code.
- Don't reinvent what already exists upstream. Prefer mjlab (GitHub, local) or other dependencies over new boilerplate in mjswan.
- Keep comments minimal and concise. Only comment the non-obvious why, not the what. When touching existing comments, strip anything redundant or self-evident.
Layout
- src/mjswan/ — package source (builder, project, scene, adapters, CLI, managers).
- src/mjswan/template/ — frontend source (Vite + React + three.js + mujoco-wasm).
- examples/ —
demo,mjlab,colab,tutorialrunnable examples. - tests/ — pytest suite.
slow-marked tests are opt-out (see below). - docs/ — zensical (MkDocs-based) site published to Read the Docs. Build with
make docs-build; serve locally withmake docs-serve.
Python workflow
Use uv instead of bare python/pip. Prefer the Makefile targets.
Tests are configured with slow opt-out: make test runs everything, but pre-commit runs pytest -m "not slow" for speed.
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 · 31 lines · 464 tokens per session scan A f4ca2992af66
mjswan AGENTS.md is an instructions file published in the GitHub repository ttktjmt/mjswan (327 stars, last pushed today), licensed Apache-2.0. It adds 464 tokens to every session, about $0.0023 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 2026-09-01.
Other instructions, from other repositories
omnisim CLAUDE.md
Claude Code instructions for omnilink-tech/omnisim, a project described as: Open-source robotics simulator for coding agents: HTTP/JSON + MCP control, Newton physics, wgpu rendering, ROS 2, and reproducible benchmarks.
omnisim AGENTS.md
AGENTS.md instructions for omnilink-tech/omnisim, covering agents.md — running omnisim as an ai coding agent, 0. for agents: read this first (60 seconds), run this first turn, first moves by task type and hard-won rules (don't relearn these).
robotics-mcp copilot-instructions.md
Copilot instructions for sandraschi/robotics-mcp: You have access to robotics orchestration: virtual robot control, ROS 2 bridge, Unity/VRChat integration, robot CRUD, and OSC control.
autonomous-os AGENTS.md
AGENTS.md instructions for autonomous-ai/autonomous-os, covering agents.md, multi-ide rules, working style, parallel work / subagents and device access rules.
claudemon CLAUDE.md
Instructions for zamarrowski/claudemon, covering ai agent guidelines, project shape, general guidelines, general rules and guards & defensive code.
unity-code-style-guide AGENTS.md
Instructions for krogh-jacobsen/unity-code-style-guide, covering agents.md — unity 6 c, project setup — edit this block, never do these — they corrupt the project, deprecated in unity 6 and if you read nothing else.