opensrc

Fetch and cache dependency source code from npm, PyPI, crates.io, and VCS references.

2|1|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/huangzida/skills --skill opensrc-huangzida
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opensrc
Source: https://github.com/huangzida/skills/tree/main/opensrc
Command: npx skills add https://github.com/huangzida/skills --skill opensrc-huangzida

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Fetches dependency source code so AI agents can read internal implementations and understand how libraries and packages work, not just their public APIs.

Core Features & Use Cases

  • Source discovery: fetches and caches exact package sources from npm, PyPI, crates.io, or VCS references for offline access.
  • Contextual reading: enables exploration of implementation details to improve reasoning about dependency behavior.
  • Use Case: debugging a library or explaining its behavior to a teammate, load the full source tree into the agent's context.

Quick Start

Ask opensrc to fetch the source for a dependency, for example, fetch the source for zod.

Frequently Asked Questions about opensrc

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

FAQPage Schema
How do I fetch dependency source code for offline analysis?

To fetch dependency source code for offline analysis, request the package source by name. The tool identifies and caches exact package sources from npm, PyPI, crates.io, or VCS references locally to enable deep dives into internal implementations.

How does caching dependency source code work for reading internal implementations?

Caching dependency source code works by cloning the exact version tag from a registry or VCS reference and storing it locally in the ~/.opensrc/ directory, exposing a path for subsequent offline reads and contextual exploration.

Can I retrieve source code from npm, PyPI, and crates.io for deep context?

Yes, you can retrieve source code from npm, PyPI, and crates.io. The tool scopes source discovery across these ecosystems and VCS references, enabling exploration of implementation details to improve reasoning about dependency behavior.

What is the best way to load a full package source tree into an agent context?

The best way to load a full package source tree into an agent context is to ask the tool to fetch the source for a specific dependency. It clones the correct version tag and exposes the local cache path for the agent to read.

Why do I need to fetch exact dependency sources instead of just reading public APIs?

You need exact dependency sources to understand how libraries and packages work internally, not just their public APIs. Fetching the full source tree allows agents to explore implementation details, debug a library, or explain its behavior to a teammate.