commit

When committing changes to this repository, follow these steps:

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "commit" with this command: npx skills add agntswrm/agent-media/agntswrm-agent-media-commit

Commit Workflow

When committing changes to this repository, follow these steps:

  1. Build and typecheck

Run both commands and ensure they pass:

pnpm build && pnpm typecheck

Do not proceed if either fails. Fix all errors first.

  1. Check what changed

git status git diff --stat

  1. Create a changeset file

IMPORTANT: The CLI package is named agent-media (NOT @agent-media/cli ).

Create .changeset/<descriptive-name>.md :


"agent-media": patch|minor|major "@agent-media/core": patch|minor|major "@agent-media/providers": patch|minor|major "@agent-media/image": patch|minor|major "@agent-media/audio": patch|minor|major "@agent-media/video": patch|minor|major

Brief description of changes

Only include packages that were actually modified. Use:

  • patch for bug fixes

  • minor for new features (backward compatible)

  • major for breaking changes

  1. Create feature branch

git checkout -b feat/<descriptive-name>

or fix/<descriptive-name> for bug fixes

  1. Stage and commit

git add <files> .changeset/<name>.md git commit -m "feat|fix: descriptive message"

  1. Push and create PR

git push -u origin <branch-name> gh pr create --title "..." --body "..."

Note: If you need to switch GitHub accounts for PR creation, check .claude.local/workflow.md for your personal account switching commands.

Important reminders

  • NEVER manually edit CHANGELOG.md - changesets auto-generates it

  • NEVER run pnpm changeset version locally

  • ALWAYS run pnpm build && pnpm typecheck before committing

  • ALWAYS sync README.md to packages/cli/README.md when README changes

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

image-remove-background

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

audio-transcribe

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

video-generate

No summary provided by upstream source.

Repository SourceNeeds Review