ソース情報
- リポジトリ
- dartsim/dart
- ソースの最終更新活動
- 2026年9月6日 16:57
- 検出された SKILL.md の言語
- 英語
- スター
- 1,202
- フォーク
- 304
インストール方法
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
ソースファイルを確認
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
メニュー
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
SOC 職業分類に基づく
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/dartsim/dart --skill dart-prコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか?SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
DART Architecture: the DART 7 multi-physics, multi-solver, multi-backend simulation pipeline, the living architecture map that draws it, and where each abstraction is owned
DART Docs Update: update docs or AI instruction visibility without code changes, or audit the docs tree for stale, duplicated, and orphaned content
DART Architecture: the DART 7 multi-physics, multi-solver, multi-backend simulation pipeline, the living architecture map that draws it, and where each abstraction is owned
| name | dart-pr |
| description | DART PR: create a branch, commit, push, and open a DART pull request |
Use this skill in Codex to run the DART dart-pr workflow. The editable
workflow source lives in .claude/commands/; this file is its generated adapter
in the shared .agents/skills/ catalog.
/dart-pr <arguments>$dart-pr <arguments>Treat the text after the skill name as $ARGUMENTS. When the workflow
references $1, $2, etc., map those to the positional values supplied by the
user.
Prepare the PR locally within the authorized task: $ARGUMENTS Before an action requiring explicit maintainer/user approval under the owner docs, verify that existing authorization covers its action, target, and scope. Ask only for missing or changed authority after completing authorized preparation.
@AGENTS.md @docs/onboarding/contributing.md @docs/onboarding/ai-reviews.md @docs/ai/verification.md @docs/onboarding/changelog.md @.github/PULL_REQUEST_TEMPLATE.md
Follow the PR-writing guidance in
docs/onboarding/contributing.md#submitting-a-pull-request and fill the compact
PR template. Keep titles plain, scoped, and outcome-focused, without agent
prefixes. Recent PRs can supply relevant context; use the current owner guidance
rather than copying their length or structure.
Draft from the final diff and the reason for it, not from the session report. Before publication, read the rendered body as a reviewer: the opening must explain the concrete problem and solution, and the rest must earn its space through review-relevant rationale or evidence. Short but generic is not enough.
For 3D structure or behavior changes (model/scene, dynamics, collision/contact,
simulation, rendering, GUI, visual examples), use dart-verify-sim. Preserve the
owner's visible media, assessed comparisons, text oracle, claim boundaries, and
reproduction evidence. Publish transient evidence with pixi run evidence-publish
as described in docs/onboarding/agent-sim-verification.md; never commit
transient evidence. A compact template does not reduce these requirements.
Inspect scope:
git status --short --branch
git diff --stat
git diff --check
Exclude unrelated dirty files unless the user explicitly includes them.
Choose the target branch and milestone:
| Target | Milestone |
|---|---|
main | DART 7.0 |
Active DART 6 LTS release-6.* | Branch-matching DART 6.x patch |
For bug fixes, use the dual-PR flow: fix the active DART 6 LTS branch first,
then cherry-pick or reapply to main.
Before every commit, run pixi run lint. Also run pixi run build for C++
or Python changes and focused tests for behavior changes.
Create or update a topic branch when needed:
git checkout -b <type>/<topic> origin/<target-branch>
Commit only intended files with a plain descriptive commit title.
Merge the latest base branch into the PR branch before any push, and follow
the base-merge and automated-review rules in docs/onboarding/ai-reviews.md
(no inline bot replies; one trigger owner and completed fix batches).
Commit and validate the immutable candidate, obtain two clean non-author
local reviews (or the evidenced trivial exception), and pass
pixi run review-gate check <candidate> before publication. Complete this
local work before asking for missing publication approval.
Verify explicit maintainer/user approval covers pushing and opening the draft
PR; ask only for missing authority. With that approval:
git push -u origin HEAD
gh pr create --draft --base <target-branch> --milestone "<milestone>" \
--title "<plain title>" --body-file <filled-template-file>
Prefer additive follow-up commits for updates to a published PR. Amend or force-push only after explicit maintainer/user approval and only when the user requests it or a clear reason exists (removing sensitive content, repairing branch history).
Invoke the dart-changelog routine for the changelog decision, entry
wording, and PR-link follow-up. If needs the PR number, keep
the follow-up changelog commit local unless existing explicit maintainer/user
approval covers its push or PR update; otherwise request only the missing
authority.
CHANGELOG.mdFollow the single Review-Fix Loop Workflow in
docs/onboarding/ai-reviews.md, reusing explicit approval for its action,
PR, and scope. Account for the initial automatic review before requesting
another; apply the strategy checkpoint and current-head readiness gates.
Monitor CI: gh pr checks <PR_NUMBER>.