research

Gather evidence from web sources, user answers, and documentation.

133|9|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/lync-cyber/CataForge --skill research-lync-cyber
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/lync-cyber/CataForge/tree/main/.cataforge/skills/research
Command: npx skills add https://github.com/lync-cyber/CataForge --skill research-lync-cyber

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires context.

What problem does it solve?

This Skill helps when information is missing, ambiguous, or too uncertain to guess. It gathers evidence from the web, user input, or existing project documents so downstream agents can proceed with traceable facts instead of assumptions.

Core Features & Use Cases

  • Web Research: Search and fetch relevant online sources for technical comparisons, best practices, API details, or competitor analysis.
  • User Interviews: Ask focused clarification questions when requirements depend on preferences, priorities, or business decisions.
  • Document Lookup: Pull relevant information from existing project docs and context to avoid duplicate research.
  • Use Case: A planning agent can use this Skill to compare implementation options, verify an API behavior, or clarify an unclear requirement before writing a final document.

Quick Start

Use the research skill in web-search mode to investigate the question and return a concise evidence-backed summary with sources and confidence levels.

Frequently Asked Questions about research

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

FAQPage Schema
How do I resolve uncertain requirements with structured research before implementation?

Resolve uncertain requirements by collecting evidence from web sources, user interviews, or existing documentation to ensure downstream decisions rely on traceable facts instead of guesswork. The research process applies structured source capture, confidence labeling, and traceable notes.

What is the best way to verify API behavior and compare technical options?

Verifying API behavior and comparing technical options is best handled through web research that fetches relevant online sources, applies confidence labeling, and returns an evidence-backed summary. This structured research ensures technical comparisons avoid assumptions.

How do I gather missing information from users when requirements are ambiguous?

Gather missing information by conducting user interviews that ask focused clarification questions when requirements depend on business decisions or preferences. This research method captures user answers to resolve ambiguity and fill information gaps.

Can I use existing project documents to avoid duplicate research?

You can avoid duplicate research by performing document lookup to pull relevant information from existing project context. This research approach extracts traceable facts from current documentation before searching external web sources.

Does structured research require confidence labeling and source capture?

Structured research requires confidence labeling and source capture to produce traceable research notes for downstream decision making. This ensures every piece of evidence gathered from web sources, user answers, or documents is verifiable.

When should I not use web search for competitor analysis?

Web search for competitor analysis is not suitable when the information gap depends on internal preferences or business decisions. In these cases, user interviews are the appropriate research method to gather focused clarification instead of external sources.