IF --nodes OR --full:

CONTEXT: Keep the skill's reference files current with n8n updates. Chains: /n8n-discover + /n8n-update-docs + /n8n-reindex.

REQUIRES: ~/.claude/n8n-config.json

STEPS:

1. READ config — confirm connected
2. PARSE $ARGUMENTS — determine scope: full | nodes | docs | index

IF --nodes OR --full:

3. RUN /n8n-discover logic:
   • TRY <baseUrl>/rest/node-types → IF fails → scan workflows for node types
   • SAVE to docs/nodes/live-nodes.json
   • MERGE new nodes into docs/nodes/node-catalog.json (add doc_url, discovered_at)
   • REBUILD docs/nodes/doc-links.json

IF --docs OR --full:

4. RUN /n8n-update-docs logic:
   • READ docs/nodes/doc-links.json for URLs
   • FETCH each doc page → extract params, credential info, options
   • UPDATE docs/nodes/node-catalog.json with fresh data
5. CHECK key n8n docs for updates:
   • https://docs.n8n.io/api/api-reference/ → new API endpoints?
   • https://docs.n8n.io/hosting/configuration/environment-variables/ → new env vars?

IF --index OR --full:

6. RUN /n8n-reindex logic:
   • SCAN all files in docs/, projects/, .n8n-track/
   • EXTRACT semantic tags per section
   • WRITE .n8n-track/index/context-map.json + line-index.json

Always:

7. UPDATE docs/nodes/node-catalog.json _last_updated timestamp

OUTPUT:

N8N Skill Evolution @ <timestamp>

  Nodes discovered: <count> (<new_count> new)
  Docs updated: <count> node references
  Index rebuilt: <entry_count> entries

  Catalog: docs/nodes/node-catalog.json
  Doc links: docs/nodes/doc-links.json
  Index: .n8n-track/index/context-map.json

Skill is current with n8n instance + docs.n8n.io

RULES:

• Never delete existing catalog entries — only add or update
• Always include doc_url for every discovered node
• Cache live-nodes.json for 24h unless --full forces refresh
• If a doc page fetch fails → web search as fallback: n8n <name> site:docs.n8n.io

$ARGUMENTS