third-party-source-catalog

Catalog and fetch third-party library source trees from GitHub repositories and local folders.

13|6|Updated Feb 19, 2025
One-click install
npx skills add https://github.com/Tairitsua/Monica --skill third-party-source-catalog
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: third-party-source-catalog
Source: https://github.com/Tairitsua/Monica/tree/main/.claude/skills/third-party-source-catalog
Command: npx skills add https://github.com/Tairitsua/Monica --skill third-party-source-catalog

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Centralizes discovery, caching, and registration of third-party library source trees so agents and CLIs can inspect dependency code without repeatedly cloning or re-downloading archives, reducing redundant network work and improving reproducible analysis.

Core Features & Use Cases

  • Cataloging: Register GitHub repositories or existing local source trees with aliases and canonical names for reuse across agents.
  • Fetch & Cache: Download tag archives or perform shallow clones with automatic fallback and maintain a managed local cache.
  • Local scanning: Recursively discover many local source roots by manifest markers or git repositories and register them for quick resolution.
  • Manifest support: Produce and respect stub and enriched manifest files to supply retrieval metadata for downstream agents.
  • Use Case: An AI code-review agent resolves a fuzzy dependency name, fetches the preferred local path from the catalog, and inspects the source without repeatedly querying GitHub.

Quick Start

Run the included source_catalog.py script to add or fetch a repository and then return the preferred local source path for inspection.

Frequently Asked Questions about third-party-source-catalog

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I cache third-party source trees for local code inspection without repeatedly cloning GitHub repositories?

You can cache third-party source trees by registering GitHub repositories or local folders in a central catalog, which downloads tag archives or shallow clones once and provides a preferred local path for subsequent code inspection.

How does local scanning discover existing source trees for dependency caching?

Local scanning recursively searches directories for manifest markers or git repositories, automatically registering discovered project roots into the catalog so agents can resolve and inspect them without manual path entry.

Do I need git and gh command-line tools to catalog third-party source trees?

Yes, you need both git and gh command-line tools available in PATH, along with a writable filesystem runtime root, to clone repositories, download tag archives, and store manifests and cached source.

What is the best way to resolve a fuzzy dependency name to a cached local source path for an AI agent?

The best way is to query the source catalog, which uses aliases and canonical names to resolve fuzzy dependency names and immediately return the preferred local source path for agent inspection.

Can I use this source catalog with existing local source folders instead of GitHub repositories?

Yes, you can register existing local source folders alongside GitHub repositories, allowing the catalog to manage both remote downloads and local directories under a unified manifest system.

Why does my AI code-review agent repeatedly download archives when inspecting dependency code?

Your agent lacks a centralized source cache, causing redundant network work; registering dependencies in a catalog fetches archives once and reuses the cached local source for reproducible analysis.