external-context

Fetch external documentation by coordinating parallel document-specialist agents.

16|3|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/sehoon787/my-claude --skill external-context-sehoon787
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: external-context
Source: https://github.com/sehoon787/my-claude/tree/main/skills/omc/external-context
Command: npx skills add https://github.com/sehoon787/my-claude --skill external-context-sehoon787

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Fetch external documentation, references, and context for a query by coordinating parallel document-specialist agents.

Core Features & Use Cases

  • Facet-based discovery: Decomposes queries into 2-5 facets to cover diverse sources.
  • Parallel research: Spawns multiple document-specialist Claude agents to gather sources simultaneously.
  • Synthesis with citations: Produces a structured, source-cited summary ready for integration.

Quick Start

Provide a topic or question to trigger the External Context Skill and fetch external documentation and citations.

Frequently Asked Questions about external-context

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

FAQPage Schema
How do I fetch external documentation and references for a research query?

Fetching external documentation works by decomposing a query into 2-5 facets and running parallel web searches across official docs, blogs, and GitHub. Multiple document-specialist agents gather sources simultaneously, returning a synthesized summary with citations.

What is parallel knowledge retrieval for technology evaluation?

Parallel knowledge retrieval decomposes a research topic into distinct facets, spawns multiple agents to search official docs and GitHub simultaneously, and synthesizes source-cited results to accelerate technology evaluation and research.

Can I use parallel web search to gather context from official docs and GitHub?

Yes, parallel web search coordinates multiple document-specialist agents to gather context from official docs, blogs, and GitHub simultaneously. It decomposes your query into facets, runs parallel searches, and returns synthesized output with sources cited.

What's the best way to research multiple facets of a technical topic simultaneously?

The best way to research multiple facets simultaneously is decomposing a query into 2-5 facets and invoking parallel Task-based document-specialist agents. Each agent searches diverse sources concurrently, producing a structured, source-cited summary.

Does parallel document search require facet decomposition to work?

Yes, explicit facet decomposition is required. The process breaks a query into 2-5 facets, then invokes parallel Task-based agents to search official docs, blogs, and GitHub, ensuring comprehensive coverage across diverse sources before synthesizing output.

When should I not use parallel research agents for knowledge gathering?

Parallel research agents are not suited for simple, single-answer queries that require no facet decomposition. The approach requires explicit decomposition into 2-5 facets and parallel Task-based agent invocation, making it excessive for straightforward documentation lookups.