local-research

Orchestrate multi-agent research over local codebase directories and generate citation-backed reports.

1|Updated Mar 24, 2026
One-click install
npx skills add https://github.com/EdgeApp/edge-dev-agents --skill local-research-edgeapp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: local-research
Source: https://github.com/EdgeApp/edge-dev-agents/tree/main/.cursor/skills/local-research
Command: npx skills add https://github.com/EdgeApp/edge-dev-agents --skill local-research-edgeapp

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill solves the difficulty of performing comprehensive, verifiable research across a large local codebase by automating the scoping, investigation, and adversarial verification of findings.

Core Features & Use Cases

  • Multi-Agent Fan-out: Decomposes complex questions into specific research angles to ensure thorough coverage.
  • Adversarial Verification: Automatically re-opens cited files to confirm claims, ensuring every finding is grounded in reality.
  • Use Case: Use this to perform a comprehensive audit of a specific module or to trace the implementation of a complex architectural pattern across the entire repository.

Quick Start

Invoke the local-research skill by providing a research question and the target directory path to initiate a deep investigation.

Frequently Asked Questions about local-research

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

FAQPage Schema
How do I perform a deep architectural audit on a local codebase?

Perform a deep architectural audit by invoking a research process that decomposes complex questions into targeted angles and adversarially verifies file-based claims to produce citation-backed reports.

What is adversarial verification for codebase analysis?

Adversarial verification for codebase analysis automatically re-opens cited files to confirm claims, ensuring every architectural finding is grounded in reality and fully traceable.

How do I trace implementation patterns across local filesystem directories?

Trace implementation patterns across local filesystem directories by orchestrating multi-agent fan-out, which scopes specific research angles to ensure thorough coverage of complex code across the entire repository.

Can I generate citation-backed documentation from existing code repositories?

Yes, you can generate citation-backed documentation from existing code repositories by running deep multi-agent research that grounds every finding in actual file contents through adversarial verification.

When should I use multi-agent research for codebase verification?

Use multi-agent research for codebase verification when facing complex code analysis, architectural audits, or documentation verification tasks that require comprehensive scoping and traceable, citation-backed findings.