init-context

Read README and linked documentation to aggregate repository context.

4|Updated Oct 12, 2021
One-click install
npx skills add https://github.com/nicolasdao/puffy-core --skill init-context
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: init-context
Source: https://github.com/nicolasdao/puffy-core/tree/main/.agents/skills/init-context
Command: npx skills add https://github.com/nicolasdao/puffy-core --skill init-context

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams reduce manual context gathering by reading the repository's README and recursively following relevant documentation links to surface the essential project context.

Core Features & Use Cases

  • Recursive documentation discovery: scans the README and follows relevant links to surface needed context.
  • Context scaffolding: builds a lightweight knowledge base from project docs to inform questions and decisions.
  • Read-only operation: never modifies source files; reads only to assemble context for user queries.

Quick Start

Provide a topic or question and let the skill load the most relevant project context by reading the README and traversing linked docs.

Frequently Asked Questions about init-context

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

FAQPage Schema
How do I load project context from README and linked documentation files?

Recursive documentation discovery works by reading the README first, then following relevant documentation links recursively to build a lightweight knowledge base for answering questions or exploring topics.

Can I automatically traverse linked docs to build a knowledge base without modifying source files?

Yes, you can traverse linked docs without modifying source files because this Skill operates in a strict read-only mode, respecting repository boundaries while assembling context for user queries.

What is the best way to answer questions about a project by aggregating context from linked docs?

The best way to answer questions by aggregating context from linked docs is to provide a topic or question, and let the Skill load the most relevant project context by reading the README and traversing linked docs.

Does context loading work across multiple linked files and record the sequence of files read?

Yes, context loading works across multiple linked files and records the sequence of files read for transparency, ensuring you can trace exactly which documentation contributed to the assembled knowledge base.

What are the limitations of using recursive link traversal for project documentation discovery?

Limitations of recursive link traversal include operating only in read-only mode, respecting repository boundaries strictly, and relying entirely on existing README links rather than generating new documentation or modifying source files.