bump
Bump version across all 5 files, land it via PR, then tag and push
Open the comprehensive testing guide and help the user test VMark features.
$ npx -y skills add xiaolai/vmark --agent claude-codeHow it fires
How this command gets triggered: by you, by Claude, or both.
/test-guideContext preview
What this command does when you run it.
Open the comprehensive testing guide and help the user test VMark features.
Open the comprehensive testing guide and help the user test VMark features.
1. Read the testing guide at `dev-docs/testing/comprehensive-testing-guide.md` (create the directory if it does not yet exist) 2. Present a summary of test categories to the user 3. If the user specifies a category, show those specific tests 4. Help track test results if requested
1. **Text Formatting** - Bold, italic, code, strikethrough, nested marks 2. **Atomic Nodes** - Footnote and math selection behavior 3. **Mode Switching** - Cursor sync between WYSIWYG and Source (F7) 4. **File Operations** - New, open, save, save as, close 5. **Auto-Save** - Automatic saving with configurable interval 6. **Document History** - Version history, revert, pruning 7. **UI Components** - Status bar, sidebar, title bar 8. **Slash Menu** - "/" trigger menu for block insertion 9. **Search** - Find and replace (Cmd+F) 10. **Focus/Typewriter** - Focus mode (F8), Typewriter mode (F9) 11. **Settings** - Themes, fonts, CJK options 12. **Unicode** - CJK text, emoji, special characters 13. **Popups** - Link, image, footnote popups 14. **Images** - Paste, drag-drop, context menu 15. **Code/Diagrams** - Code blocks, LaTeX, Mermaid
Ask the user which category they want to test, then: 1. Show the relevant test cases from the guide 2. Help them execute tests if app is running (use Tauri MCP) 3. Record results
The Plain-Text Workspace Where Humans and AI Collaborate Free. Local-first. Format-aware. VMark is the plain-text workspace where humans and AI collaborate.
Repo: xiaolai/vmark
Bump version across all 5 files, land it via PR, then tag and push
Run the gated, agent-driven workflow end-to-end.
End-to-end GitHub issue resolver — fetch, classify, fix, audit, PR
Fix issues properly - no patches, no shortcuts, no regressions
Review and merge open PRs — sequential, safe, with rebase handling
Remove failed/canceled GitHub Actions runs and stale remote branches