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/dartsim/dart/agents-mdgit clone --depth 1 https://github.com/dartsim/dartWrote 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/dartsim/dart/agents-md)<a href="https://agentmods.dev/instructions/dartsim/dart/agents-md"><img src="https://agentmods.dev/badge/instructions/dartsim/dart/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 | $0.02156 | $0.02156 |
| Opus 5 | $0.01078 | $0.01078 |
| Sonnet 5 | $0.00431 | $0.00431 |
| Haiku 4.5 | $0.00216 | $0.00216 |
Grade A, and why
dart 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 today.
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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Guidelines for DART
This file is a pointer board for agents working in this repository. Keep it concise and expand other documents instead.
Project Overview
WHAT: C++23 physics engine for robotics simulation with Python bindings (dartpy)
WHY: Research-grade dynamics for robotics, animation, and machine learning
HOW: Build/test with pixi run tasks; format with pixi run lint before commits
Quick Commands
pixi run test-all # Lint + build + all tests (fails fast)
pixi run lint # Format code/docs (auto-fixes)
pixi run build # Build C++ and Python
pixi run test-unit # Unit tests only (faster)
pixi run test-py # Python tests only
pixi run check-lint # Check formatting without fixing
pixi run ai-doctor # Read-only agent/setup diagnosis
pixi run check-ai-infra # AI discovery, drift, hook, and scenario checks
Success: "All tests passed!"
If this fails, see docs/onboarding/building.md and
docs/onboarding/testing.md for local troubleshooting; docs/onboarding/ci-cd.md
covers hosted CI.
Context Loading (IMPORTANT)
Agents MUST load relevant docs before starting work. Start every session by
reading docs/ai/principles.md, then load the task-specific docs below. Tools
that support @file references may use them, but the paths below are the
portable source of truth.
| Task Type | Load These Files |
|---|---|
| Any task | This file (loaded first by every tool chain), docs/ai/principles.md |
| Project direction | docs/ai/north-star.md |
| Building | docs/onboarding/building.md |
| Testing | docs/onboarding/testing.md |
| Model / simulation / visual verification | docs/onboarding/agent-sim-verification.md, docs/ai/verification.md; use /dart-verify-sim (Claude Code) or $dart-verify-sim (Codex); OpenCode loads the same skill from .claude/skills/dart-verify-sim/ |
| Contributing | docs/onboarding/contributing.md, CONTRIBUTING.md |
| Code style | docs/onboarding/code-style.md |
| Docs structure / placement | docs/README.md, docs/information-architecture.md, docs/AGENTS.md |
| Architecture | docs/onboarding/architecture.md, docs/onboarding/README.md |
| DART 7 architecture vision | docs/readthedocs/architecture.md (multi-physics/solver/backend one-paper), docs/design/simulation_solver_architecture.md |
| Architecture hardening / work packets | docs/design/dart7_architecture_assessment.md (verified findings + standing rule), docs/plans/solver-family-intake.md, docs/ai/orchestration.md |
| Public API work | docs/onboarding/api-boundaries.md |
| Theory/research foundations | docs/background/README.md |
| CI/CD issues | docs/onboarding/ci-cd.md |
| Python bindings | docs/onboarding/python-bindings.md |
| Model loading | docs/onboarding/io-parsing.md |
| Build system | docs/onboarding/build-system.md |
| Profiling / performance | docs/onboarding/profiling.md |
| AI docs / instruction edits | docs/ai/README.md read order: docs/ai/principles.md, docs/ai/components.md, docs/ai/verification.md; docs/ai/terminology.md when naming things |
| Tool compatibility / adapters | docs/onboarding/ai-tools.md, docs/ai/components.md |
| Planning | docs/ai/principles.md, docs/ai/north-star.md, docs/plans/README.md, docs/plans/dashboard.md, docs/plans/north-star-roadmap.md, docs/ai/verification.md |
| PR reviews | docs/onboarding/ai-reviews.md (automated review handling rules) |
| Release work | docs/onboarding/release-management.md |
| Changelog work | docs/onboarding/changelog.md, docs/onboarding/release-roadmap.md, docs/onboarding/release-management.md; use /dart-changelog or $dart-changelog for changelog decisions and entries |
| Dev tasks | docs/dev_tasks/README.md (when to create, cleanup rules) |
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.
- today First seen · 123 lines · 2,156 tokens per session scan A c89048ab5f90
dart AGENTS.md is an instructions file published in the GitHub repository dartsim/dart (1,200 stars, last pushed today), licensed BSD-2-Clause. It adds 2,156 tokens to every session, about $0.0108 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 2026-09-03.
Other instructions, from other repositories
missionos AGENTS.md
Instructions for pome223/missionos, covering agent instructions, documentation layers, claim boundaries, public safety and pull request verification.
claude-ros2-skills CLAUDE.md
Claude Code instructions for Leehyunbin0131/claude-ros2-skills, covering read this first — ros 2 verification protocol, establish before writing (no doc can tell you these) and done means it ran.
CFDesktop AGENTS.md
Instructions for Awesome-Embedded-Learning-Studio/CFDesktop, covering agents.md — codex cli entry point, at a glance, read first and codex-specific.
CFDesktop CLAUDE.md
Instructions for Awesome-Embedded-Learning-Studio/CFDesktop, covering claude.md — claude code entry point and claude-specific.
sage-unreal-mcp AGENTS.md
AGENTS.md instructions for alemdarlabs/sage-unreal-mcp, covering sage unreal mcp - agent guide, communication, skill routing, product state and canonical layout.
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).