external-context

Fetch external documentation and synthesize cited briefings from web search results.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/INNERJOINT/HarnessSkills --skill external-context-innerjoint
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: external-context
Source: https://github.com/INNERJOINT/HarnessSkills/tree/main/skills/external-context
Command: npx skills add https://github.com/INNERJOINT/HarnessSkills --skill external-context-innerjoint

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates the discovery and synthesis of external documentation and web references for a given query by decomposing complex information needs into focused search facets and aggregating authoritative sources into a single, cited briefing.

Core Features & Use Cases

  • Facet Decomposition: Breaks an input query into 2-5 independent search facets to cover different angles of a topic.
  • Parallel Document Specialists: Launches parallel specialist agents to perform targeted web search and fetch operations, each citing discovered sources with URLs.
  • Synthesis & Citation: Produces a structured output with key findings, facet-level details, and a consolidated sources list for developer research, comparative evaluations, or documentation lookup.
  • Use Case: Compare two database ORMs by running facet-based searches for features, performance guidance, and migration patterns, then receive a consolidated, cited summary.

Quick Start

Use the external-context skill to retrieve authoritative documentation and a cited summary for "Compare Prisma vs Drizzle ORM for PostgreSQL".

Frequently Asked Questions about external-context

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

FAQPage Schema
How do I gather authoritative external documentation for a complex technical query?

To gather external documentation, the system decomposes a query into 2-5 independent search facets and runs parallel specialist agents to fetch and synthesize web results into a concise, cited briefing.

What is facet decomposition for web search and how does it work?

Facet decomposition breaks an input query into 2-5 focused search angles, allowing parallel specialist agents to cover different aspects of a topic independently before aggregating results into a synthesized briefing.

Can I compare two technologies using parallel web search and get cited sources?

Yes, you can compare technologies by running facet-based searches for features, performance, and migration patterns, receiving a consolidated summary with explicit source citations and URLs.

Does external documentation synthesis require parallel agent orchestration?

Yes, external documentation synthesis requires parallel agent orchestration to launch specialist agents that perform targeted web search and fetch operations simultaneously across multiple query facets.

What's the best way to structure a query for comparative technology analysis?

Structure a comparative technology query by specifying the technologies and aspects to evaluate, such as "Compare Prisma vs Drizzle ORM for PostgreSQL," enabling facet decomposition into targeted feature and performance searches.

When should I not use a multi-agent approach for web search synthesis?

Avoid multi-agent web search synthesis for simple, single-fact lookups where one search query suffices, as facet decomposition and parallel orchestration add overhead better suited for complex, multi-angle research tasks.