tokyo-megacorp/enrich
skills/enrich/SKILL.md
Re-fetch and enrich thin Figma frames that have shallow data (1 node, generic description like "dark mode screen", or description under 30 chars). Uses REST API to fetch full node JSON and regenerates frame.md for each thin frame. Use when the user runs /figma-differ:enrich, says "enrich thin frames", "improve frame descriptions", or "deep-fetch all frames".
development
Updated Apr 23, 2026
$ install --global
skillsauthnpx skillsauth add tokyo-megacorp/figma-differ enrichInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 23, 2026, 7:51 PM154.0s1 file scanned
SKILL.md
- name:
- enrich
- description:
- >
- frame. Use when the user runs /figma-differ:
- enrich, says "enrich thin frames",
- argument-hint:
- <figma-file-url-or-fileKey>
Enrich Thin Frames
1. Parse arguments
Extract fileKey from $ARGUMENTS:
- If it looks like a Figma URL, extract the fileKey segment:
https://www.figma.com/design/<fileKey>/... - If it's a bare fileKey (no slashes), use it directly
2. Check prerequisites
bash $CLAUDE_PLUGIN_ROOT/scripts/auth.sh status
If it fails, tell the user to run bash $CLAUDE_PLUGIN_ROOT/scripts/auth.sh set and stop.
Verify the fileKey has local data:
ls ~/.figma-differ/<fileKey>/
If missing, tell the user to run /figma-differ:track <url> first.
3. Orchestration
TaskCreate("Scan thin frames", activeForm: "Detecting frames with shallow data...")
TaskCreate("Re-fetch & enrich", activeForm: "Fetching full node JSON for thin frames...")
TaskCreate("Rebuild frame docs", activeForm: "Regenerating frame.md documents...")
TaskCreate("Update search index", activeForm: "Re-indexing enriched frames in QMD...")
Task lifecycle:
- Create ALL tasks upfront (pending)
- Before each phase:
TaskUpdate(taskId, status: "in_progress") - Dispatch
Agent(model: "haiku")— agent reports only counts, never raw output - After agent returns:
TaskUpdate(taskId, status: "completed") - Every task must be completed or deleted before final summary
Data discipline
- NEVER keep raw API responses or JSON in conversation context
- ALWAYS write intermediate data to temp files
- Subagent prompts MUST include: "Write results to <path>. Report only counts and file paths."
4. Run enrichment
Dispatch a haiku agent to run the script:
Prompt: Run enrich-thin-frames.sh for fileKey <fileKey>.
Command: bash $CLAUDE_PLUGIN_ROOT/scripts/enrich-thin-frames.sh <fileKey>
Write stdout to /tmp/enrich-<fileKey>-result.txt.
Report only: total frames scanned, enriched count, skipped count, QMD status.
5. Report
Enrich complete: <fileKey>
Frames scanned: N
Enriched: M (were thin — re-fetched and frame.md regenerated)
Already rich: K (skipped)
QMD index: updated | unchanged
Run /figma-differ:search "query" to verify enriched frames are now findable.
Related Skills
tokyo-megacorp/track
testing
VerifiedTrustedCommunity
Subscribe to a Figma file for automatic syncing and semantic search. Adds the file to tracked.json, runs initial index + snapshot-all + frame.md generation, and initializes the QMD search collection. Use when the user runs /figma-differ:track or says "track this Figma file", "subscribe to Figma", "watch this design file", or "add to tracked files".
SKILL.mdUpdated Apr 23, 2026
tokyo-megacorp/sync
documentation
VerifiedTrustedCommunity
Refresh snapshots and search index for tracked Figma files. Fetches current state from Figma, generates frame.md documents, and updates the QMD search index. Use when the user runs /figma-differ:sync or says "sync Figma", "refresh snapshots", "update the Figma index", or "re-sync tracked files".
SKILL.mdUpdated Apr 23, 2026
tokyo-megacorp/snapshot
tools
VerifiedTrustedCommunity
Takes a snapshot of a Figma node — fetches its JSON structure and PNG screenshot and stores both to ~/.figma-differ/ for later diffing. Use when the user runs /figma-differ:snapshot with a Figma URL, or says "snapshot this Figma frame", "save a Figma baseline", or "take a Figma snapshot".
SKILL.mdUpdated Apr 23, 2026
tokyo-megacorp/snapshot-all
development
VerifiedTrustedCommunity
Bulk snapshots every frame in a Figma file — fetches all node JSONs, exports PNGs, and stores comments. Uses a single API call for the tree and batched image exports. Use when the user runs /figma-differ:snapshot-all with a Figma file URL, or says "snapshot all frames", "bulk snapshot", "baseline the whole file", or "snapshot everything in this Figma file".
SKILL.mdUpdated Apr 23, 2026