DART is an open-source C++23 physics engine that simulates the movement and interactions of articulated rigid-body systems for robotics, animation, and machine learning. Researchers and developers use it for kinematics, dynamics, collision handling, constraints, and loading robot models, with C++ and Python interfaces. The catalogue add-ons support workflows built around this engine.
Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/dartsim/dartnpx agentmods add skills/dartsim/dart/dart-prWrote 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/skills/dartsim/dart/dart-pr)<a href="https://agentmods.dev/skills/dartsim/dart/dart-pr"><img src="https://agentmods.dev/badge/skills/dartsim/dart/dart-pr/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/skills/dartsim/dart/dart-pr"><img src="https://agentmods.dev/badge/skills/dartsim/dart/dart-pr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00022 | $0.01839 |
| Opus 5.5 | $0.00009 | $0.00736 |
| Sonnet 5.5 | $0.00004 | $0.00368 |
| Haiku 4.5 | $0.00002 | $0.00184 |
Grade A, and why
dart-pr 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 23d 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 — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dart-pr
Use this skill to run the DART dart-pr workflow. The editable workflow
source lives in .claude/commands/; this generated skill is its entrypoint for
Codex and Muse.
Invocation
- Claude Code:
/dart-pr <arguments> - Codex:
$dart-pr <arguments> - Muse:
/dart-pr <arguments>
Treat the text after the skill name, or the user's request when the skill is
loaded on demand, as $ARGUMENTS. When the workflow references $1, $2,
etc., map those to the positional values supplied by the user.
Command Body
Prepare or open a DART pull request after explicit maintainer/user approval: $ARGUMENTS
Required Reading
@AGENTS.md @docs/onboarding/contributing.md @docs/onboarding/release-management.md @docs/onboarding/ai-tools.md @docs/onboarding/changelog.md @.github/PULL_REQUEST_TEMPLATE.md
Recent PR Patterns
When project context is unclear, inspect recently merged PRs for examples; the current PR-description owner below takes precedence over older bodies:
gh pr list --repo dartsim/dart --state merged --base <target-branch> --limit 10 \
--json number,title,body,mergedAt
Use these practices:
- Keep titles plain, scoped, and outcome-focused. Do not add agent prefixes.
- Describe private plans in prose or link public PRs/issues. Keep private, local, and machine-specific paths out of the title and body.
- Follow
docs/onboarding/contributing.md§ "PR Descriptions" for the Summary and Testing default, optional material context, evidence formats, and collapsed detail. This is the PR-description owner for agents and contributors. Write each paragraph and list item of the body on one line: GitHub renders newlines as line breaks, so never wrap PR text at a fixed width. - For CI, performance, or infrastructure work, include evidence that supports the outcome and relevant verification gaps.
- For model/scene, dynamics, collision/contact, simulation, rendering, mesh,
texture, GUI, or visual-example changes, use
dart-verify-sim: report the text correctness oracle and include assessed, claim-tied Filament/debug-overlay evidence when applicable (an image alone is not correctness proof):- Prefer an existing headless example path such as
--headless,--frames,--width,--height, and--shotover manual screenshots. - Follow
docs/ai/verification.md§ "Simulation Verification Route" for before/after 3D highlights, matched captures, and compute-only plots. - Inspect the images yourself and retain reproduction commands and environment details in collapsed evidence details. Keep limiting caveats and failed or uncertain results prominent.
- Use
pixi run evidence-publishto render the section without network access. Followdocs/ai/verification.md§ "Visual Verification (Headless Capture)" for publication after explicit maintainer/user approval. Attachments use the same store as drag-and-drop, never release assets or tags;--body-filereplaces the whole PR body. Keep transient PR evidence out of Git; commit visual files only as durable documentation, fixtures, or source assets. - If no headless path exists, either add a narrowly scoped capture mode when it fits the example or document why visual comparison is not practical.
- Prefer an existing headless example path such as
- Link relevant PRs, issues, backports, and follow-ups when useful to the review.
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.
- 23d ago Changed · +2 lines c220dab8473b
- 25d ago Changed · -11 lines f5c2c1f83969
- 26d ago Changed · +4 lines 1b830cc83376
- 28d ago Changed · -33 lines e6392614c2cf
- 29d ago Changed · -3 lines 17591ed6d406
- 1mo ago Changed · +1 lines 6956aa9bc0cf
- 1mo ago Changed · +93 lines 540c0666c253
- 2mo ago Changed · -14 lines c949530ed8fa
- 2mo ago Changed · -17 lines 1bdd7b561cf7
- 2mo ago First seen · 154 lines · 22 tokens per session scan A 0ea55b614c60
dart-pr is a skill published in the GitHub repository dartsim/dart (1,223 stars, last pushed yesterday), licensed BSD-2-Clause. It adds 22 tokens to every session and 1,839 once invoked, about $0.0001 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 2026-09-01.
dart-pr compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| dart-prdartsim/dart | Skill | 1.2k | yesterday | BSD-2-Clause | A | 22 | Claude Code |
| review-deltatirth8205/code-review-graph | Skill | 33k | 8d ago | MIT | A | 24 | Claude Code |
| hero-diff-highlightUnclecheng-li/AI_Animation | Skill | 1.7k | 11d ago | MIT | A | 0 | Claude Code, Codex |
| lean-reviewSnowflake-Labs/cocoplus | Skill | 727 | 2d ago | MIT | A | 44 | Claude Code |
| git-pr-workflows-git-workflowrmyndharis/antigravity-skills | Skill | 1.9k | 23d ago | MIT | A | 37 | Claude Code |
| review-gatemajiayu000/spellbook | Skill | 290 | 2d ago | MIT | A | 95 | Claude Code, Codex |
| dyl-reviewcursor/plugins | Skill | 15k | yesterday | none | A | 80 | Claude Code, Cursor |