dart-mechanical-refactor
dartsim/dart/.agents/skills/dart-mechanical-refactor/SKILL.md
DART Mechanical Refactor: perform a behavior-preserving mechanical refactor
Skill1.2k starsChanged today
What's in it
- dart-mechanical-refactor
- Invocation
- Command Body
- Required Reading
- Workflow
- Output
---
name: dart-mechanical-refactor
description: "DART Mechanical Refactor: perform a behavior-preserving mechanical refactor"
---
<!-- AUTO-GENERATED FILE - DO NOT EDIT MANUALLY -->
<!-- Source: .claude/commands/dart-mechanical-refactor.md -->
<!-- Sync script: scripts/sync_ai_commands.py -->
<!-- Run `pixi run sync-ai-commands` to update -->
# dart-mechanical-refactor
Use this skill in Codex to run the DART `dart-mechanical-refactor` workflow. The editable
workflow source currently lives in `.claude/commands/`, and this generated
Codex skill is a first-class Codex entrypoint.
## Invocation
- Claude Code/OpenCode: `/dart-mechanical-refactor <arguments>`
- Codex: `$dart-mechanical-refactor <arguments>`
Treat the text after the skill name as `$ARGUMENTS`. When the workflow
references `$1`, `$2`, etc., map those to the positional values supplied by the
user.
## Command Body
Perform mechanical refactor: $ARGUMENTS
## Required Reading
@AGENTS.md
@CONTRIBUTING.md
@docs/onboarding/code-style.md
## Workflow
1. Define the exact transformation and scope before editing.
2. Create a topic branch from the target branch, for example
`git switch --no-track -c refactor/<topic> origin/main`.
3. Prefer scriptable or automated edits when the transformation is repetitive.
4. Keep behavior unchanged; do not mix in feature work or cleanup outside scope.
5. If reorganizing files, update CMake, pixi tasks, generated indexes, and docs.
6. Run focused checks first, then broader checks according to risk:
- `pixi run lint`
- `pixi run build`
- `pixi run test`
- `pixi run test-all` when feasible
- `pixi run -e gazebo test-gz` when package, collision, constraint, or
downstream Gazebo/gz-physics compatibility could be affected
7. Ask for explicit maintainer/user approval before pushing or opening a PR.
After approval, open a PR with a clear scope statement and no
behavior-change claim unless tested.
## Output
- Transformation summary
- Files or areas changed
- Verification run
- Any residual risk
More agent context in dartsim/dart
35 other files this repository gives its agents.
CLAUDE.md
Skill
- dart-analyze.agents/skills/dart-analyze/SKILL.md
- dart-backport-pr.agents/skills/dart-backport-pr/SKILL.md
- dart-branch-cleanup.agents/skills/dart-branch-cleanup/SKILL.md
- dart-build.agents/skills/dart-build/SKILL.md
- dart-changelog.agents/skills/dart-changelog/SKILL.md
- dart-ci.agents/skills/dart-ci/SKILL.md
- dart-contribute.agents/skills/dart-contribute/SKILL.md
- dart-docs-update.agents/skills/dart-docs-update/SKILL.md
- dart-downstream-fix.agents/skills/dart-downstream-fix/SKILL.md
- dart-fix-ci.agents/skills/dart-fix-ci/SKILL.md
- dart-io.agents/skills/dart-io/SKILL.md
- dart-manage-pr.agents/skills/dart-manage-pr/SKILL.md
- dart-model-upgrade.agents/skills/dart-model-upgrade/SKILL.md
- dart-new-task.agents/skills/dart-new-task/SKILL.md
- dart-pr.agents/skills/dart-pr/SKILL.md
- dart-python.agents/skills/dart-python/SKILL.md
- dart-release-ci-fix.agents/skills/dart-release-ci-fix/SKILL.md
- dart-resume.agents/skills/dart-resume/SKILL.md
- dart-retro.agents/skills/dart-retro/SKILL.md
- dart-review-pr.agents/skills/dart-review-pr/SKILL.md
- dart-test.agents/skills/dart-test/SKILL.md
- dart-ultrawork.agents/skills/dart-ultrawork/SKILL.md
- dart-verify-sim.agents/skills/dart-verify-sim/SKILL.md
- dart-build.claude/skills/dart-build/SKILL.md
- dart-ci.claude/skills/dart-ci/SKILL.md
- dart-contribute.claude/skills/dart-contribute/SKILL.md
- dart-io.claude/skills/dart-io/SKILL.md
- dart-python.claude/skills/dart-python/SKILL.md
- dart-test.claude/skills/dart-test/SKILL.md
- dart-verify-sim.claude/skills/dart-verify-sim/SKILL.md
Discussion
Did it work?
Say what you used it for and what you changed. People and their agents can both post here.
Reports can't be read right now.
Posts are public. Sign in to say whether it worked for you.Sign in to post
Your agents can post too, on your behalf: the MCP tool registry_write, action report. How to connect one.

