From plugin-dev-sync
Use when user wants to This skill should be used when the user asks to "add a resource", "update resources", "check resource status", "add a documentation source", "add a Twitter account to track", or needs to manage the plugin-dev-sync resource list. Provides guidance for maintaining the master resources.json that drives the discovery pipeline. This skill should be used when the user asks to "add a resource", "update resources", "check resource status", "add a documentation source", "add a Twitter account to track", or needs to manage the plugin-dev-sync resource list. Provides guidance for maintaining the master resources.json that drives the discovery pipeline.
npx claudepluginhub karimstekelenburg/k3m-marketplace --plugin plugin-dev-syncThis skill uses the workspace's default tool permissions.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Guides slash command development for Claude Code: structure, YAML frontmatter, dynamic arguments, bash execution, user interactions, organization, and best practices.
Maintain the master resource list at data/resources.json that drives plugin-dev-sync's
discovery and sync pipeline. Every resource the system crawls must be registered here.
The resource file lives at ${CLAUDE_PLUGIN_ROOT}/data/resources.json.
Each resource entry has these fields:
| Field | Required | Description |
|---|---|---|
id | yes | Unique kebab-case identifier |
url | yes | Fetchable URL |
name | yes | Human-readable label |
category | yes | One of: official-docs, github, social, community |
type | yes | One of: page, sitemap-root, repo, releases, directory, twitter, rss, api |
notes | no | Context for crawlers — what to look for |
status | yes | active or inactive |
data/resources.jsoncurl -sI <url>id following the pattern: <category>-<short-name> (e.g., docs-hooks, twitter-karpathy)resources arrayuv run ${CLAUDE_PLUGIN_ROOT}/scripts/fetch-resources.py --id <new-id>Prefer setting "status": "inactive" over deleting, to preserve history.
Only delete entries that were added by mistake.
Run the fetch script to check all resources are reachable:
uv run ${CLAUDE_PLUGIN_ROOT}/scripts/fetch-resources.py
Review data/fetch-report.json for failures. Common issues:
The fetch script at scripts/fetch-resources.py supports filtering:
# Fetch everything
uv run scripts/fetch-resources.py
# Fetch one category
uv run scripts/fetch-resources.py --category official-docs
# Fetch one resource
uv run scripts/fetch-resources.py --id docs-claude-code-hooks
Results are cached in data/cache/<id>.txt and summarized in data/fetch-report.json.