external-research

Guide AI agents through targeted web research with source citations.

Updated Feb 3, 2026
One-click install
npx skills add https://github.com/qdouble/stack-contracts --skill external-research
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: external-research
Source: https://github.com/qdouble/stack-contracts/tree/main/.agent/skills/external-research
Command: npx skills add https://github.com/qdouble/stack-contracts --skill external-research

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a structured approach for AI agents to conduct external web research, ensuring they gather necessary information before planning or executing tasks, especially when dealing with unfamiliar APIs, platform-specific issues, or security-sensitive features.

Core Features & Use Cases

  • Guided Research Triggers: Defines clear conditions for when research is necessary (e.g., unfamiliar APIs, error codes) and when it's not (e.g., clear codebase answers).
  • Research-First Planning Pattern: Outlines a systematic process: identify unknowns, research, synthesize findings, and cite them in plans.
  • Source Quality Hierarchy & Citation: Establishes trust levels for different sources and provides standardized formats for citing research in plans, code comments, and collaboration files.
  • Use Case: When tasked with implementing a new authentication flow on a specific mobile platform, the agent uses this skill to research platform-specific security best practices and API nuances before generating the implementation plan.

Quick Start

Use the external-research skill to find the recommended way to handle background task cancellation on watchOS.

Frequently Asked Questions about external-research

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

FAQPage Schema
How do I research unfamiliar APIs before implementing platform-specific code?

When dealing with unfamiliar technical domains, platform-specific challenges, or security-sensitive features, you should trigger external research before planning or executing any related coding tasks.

What is the best way to structure web research findings for AI planning?

The best way to structure web research findings is by using a research-first planning pattern that synthesizes information and cites sources using a standardized format within plans and code comments.

When should an AI agent avoid web search and rely on the existing codebase?

An AI agent should avoid web search and rely on the existing codebase when clear answers are already present locally, avoiding unnecessary research for known APIs or internal logic.

How do you prioritize information sources when researching technical documentation?

Prioritize information sources by applying a source quality hierarchy that establishes trust levels, ensuring that technical documentation and API research are accurately cited and integrated.

Can I use this approach to handle security-sensitive features on mobile platforms?

Yes, you can use this approach for mobile platforms by researching platform-specific security best practices and API nuances before generating the final implementation plan.