Menu
Use this to capture screen context.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
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