external-context

Decompose queries into search facets and synthesize cited web documentation.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/nichobbs/lyric-lang --skill external-context-nichobbs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: external-context
Source: https://github.com/nichobbs/lyric-lang/tree/main/.claude/skills/external-context
Command: npx skills add https://github.com/nichobbs/lyric-lang --skill external-context-nichobbs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It reduces the time and uncertainty of researching unfamiliar topics by orchestrating parallel web lookups and compiling results into a single, cited brief.

Core Features & Use Cases

  • Facet-based research decomposition: Breaks a query into 2–5 independent search facets so different angles are covered efficiently.
  • Parallel document-specialist lookups: Spawns up to 5 concurrent agents to gather official docs, examples, and references per facet.
  • Cited synthesis output: Produces a structured summary with key findings, per-facet detailed results, and a consolidated sources list.

Quick Start

Invoke it with /oh-my-claudecode:external-context and your question or topic to receive a synthesized, URL-cited research brief.

Frequently Asked Questions about external-context

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

FAQPage Schema
How do I gather sourced web research for an unfamiliar coding architecture question?

You can gather sourced web research by decomposing a coding architecture question into multiple independent search facets and spawning parallel agents to collect official documentation, examples, and references. This approach compiles a cited brief with key findings and consolidated URLs from multiple concurrent perspectives.

What is the best way to compare libraries using documentation lookup from multiple angles?

The best way to compare libraries is through query decomposition into 2–5 independent search facets, enabling parallel document-specialist lookups across different angles. This concurrent perspective gathering produces a structured summary with detailed per-facet results and a final list of source URLs.

How do I get cited best-practice lookups from web search for an unfamiliar technical topic?

You get cited best-practice lookups by orchestrating parallel web lookups that spawn up to five concurrent agents to gather official docs and references per facet. The agents synthesize results into a structured summary with key findings, per-facet details, and a consolidated sources list.

Can I use parallel agents to research multiple independent search facets simultaneously?

Yes, you can use parallel agents to research multiple independent search facets simultaneously, with a maximum of five concurrent agents. Each agent acts as a document specialist gathering official documentation, examples, and references for its assigned facet, which are then compiled into a single cited synthesis output.

Does parallel external research work for documentation-driven investigations with multiple concurrent perspectives?

Yes, parallel external research works effectively for documentation-driven investigations by decomposing queries into 2–5 independent search facets and collecting documentation from the web. It applies to best-practice lookups, library comparisons, and architecture questions that benefit from multiple concurrent perspectives.

What are the limitations of using facet decomposition for external web research?

Facet decomposition for external web research is limited to a maximum of five concurrent parallel agents and decomposes queries into 2–5 independent search facets. It is specifically designed for coding, architecture, library comparisons, best-practice lookups, and documentation-driven investigations rather than general browsing.