Menu
Use this to capture screen context.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Check papers/ structure for completeness and accuracy — every PDF has text extraction, every paper is indexed in PAPER_INDEX.md and summarized in PAPER_SUMMARIES.md, summaries are factually accurate. Prompts the user for discrepancies.
Use when you want to clean up accumulated git worktrees in a repo. Lists all worktrees with status, verifies data symlinks, ensures code and docs are committed, copies non-symlinked data files to main worktree, asks which to merge or keep, runs post-merge audits on new docs/papers, and removes merged worktrees.
Use after creating or modifying decision documents in docs/ to keep DOCS_INDEX.md and DOCS_SUMMARY.md consistent. Also use as a standalone audit to find coverage gaps, broken links, and ADR format issues. Works across any repo with a docs/ directory. Run alongside updating-noridocs when finishing a branch that includes decision docs.
IMMEDIATELY USE THIS SKILL when creating or develop anything and before writing code or implementation plans - refines rough ideas into fully-formed designs through structured Socratic questioning, alternative exploration, and incremental validation
End a research session — runs update-docs to checkpoint all progress, then commits and pushes. Use update-docs for mid-session checkpoints without ending the session.
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
| name | Taking and Analyzing Screenshots |
| description | Use this to capture screen context. |
You CAN take screenshots by combining the Bash tool with platform-specific screenshot commands. Screenshots are saved as image files, then loaded into your context using the Read tool for visual analysis.
Use this skill when I ask you to:
| Platform | Command | Interactive Selection |
|---|---|---|
| macOS | screencapture | -i flag (area selection) |
| Linux | gnome-screenshot, scrot, or import | -a or -s flag |
Standard workflow:
uname -s/tmp/screenshot_$(date +%s).pnguname -s
Darwin for macOSLinux for LinuxmacOS: screencapture is always available (built-in)
Linux: Check in priority order:
which gnome-screenshot || which scrot || which import || echo "none"
Priority order (best compatibility):
gnome-screenshot - works on both X11 and Waylandscrot - lightweight, X11 onlyimport - part of ImageMagickUse timestamped filename to avoid conflicts:
macOS:
screencapture -i /tmp/screenshot_$(date +%s).png
-i enables interactive area selectionLinux with gnome-screenshot:
gnome-screenshot -af /tmp/screenshot_$(date +%s).png
-a for area selection-f specifies filenameLinux with scrot:
scrot -s /tmp/screenshot_$(date +%s).png
-s enables selection modeLinux with import:
import /tmp/screenshot_$(date +%s).png
Read tool: file_path="/tmp/screenshot_12345.png"
The Read tool displays images visually. You'll see the screenshot and can analyze it.
Once loaded, you can:
rm /tmp/screenshot_12345.png
Only remove if I won't need the file again.
If no screenshot tool is available on Linux:
Inform me which tool is missing
Suggest installation:
sudo apt install gnome-screenshotsudo dnf install gnome-screenshotsudo pacman -S gnome-screenshotsudo apt install imagemagick (or equivalent)Alternative: Ask user to manually take screenshot and provide path
Reality: You CAN via Bash + screenshot CLI tools
Reality: The screenshot file must be loaded with Read tool to see it
Reality: Always use absolute paths (/tmp/...) for Read tool
Reality: Must detect which tool is installed before attempting capture
User: "Take a screenshot and help me debug this UI bug"
1. Check platform:
uname -s → Linux
2. Check available tools:
which gnome-screenshot → /usr/bin/gnome-screenshot
3. Capture screenshot:
gnome-screenshot -af /tmp/screenshot_1729012345.png
→ User selects area, file saved
4. Load into context:
Read: file_path="/tmp/screenshot_1729012345.png"
→ Image displays visually
5. Analyze:
"I can see the button alignment is off. The 'Submit' button
is 5px lower than the 'Cancel' button..."
6. Optional cleanup:
rm /tmp/screenshot_1729012345.png