research

Scan dependency manifests, docs, and git history to produce .godmode/research.md.

26|8|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/arbazkhan971/godmode --skill research-arbazkhan971
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/arbazkhan971/godmode/tree/main/skills/research
Command: npx skills add https://github.com/arbazkhan971/godmode --skill research-arbazkhan971

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? This Skill closes the gap between local codebase analysis and external prior art by systematically collecting existing patterns, relevant dependencies, docs, and prior attempts so downstream planning can build on real context instead of reinventing solutions.

## Core Features & Use Cases

  • Explorer-driven code mapping: Dispatches a read-only explorer subagent to map files, symbols, and abstractions related to the feature.
  • Ecosystem & history scan: Searches dependency manifests, docs, ADRs/RFCs, and git history to identify reusable packages, constraints, and prior attempts.
  • Concise synthesis and commit: Produces a compact .godmode/research.md (under 60 lines) with relevance tags, appends a TSV log entry, and commits the research before think runs.
  • Use Case: When adding a cross-cutting feature or integrating an external library, use this Skill to avoid proposing greenfield designs and to anchor proposals in existing patterns.

### Quick Start Use the research skill to gather prior art for the proposed "add user authentication" feature and write .godmode/research.md.

Frequently Asked Questions about research

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

FAQPage Schema
How do I gather prior art and external context before implementing a feature?

To gather prior art for a proposed feature, you can scan dependency manifests, docs, and git history to map existing patterns, constraints, and prior attempts. This produces a concise research.md file with relevance tags.

When should I perform a dependency scan and codebase scan for feature work?

You should scan dependencies and the codebase when adding cross-cutting features or integrating external libraries. This ensures you anchor proposals in existing patterns rather than reinventing solutions.

How does an explorer agent map existing code patterns for new development?

An explorer agent maps files, symbols, and abstractions related to a proposed feature in a read-only pass. It searches ADRs, RFCs, and git history to identify reusable packages and prior attempts.

What is the best way to document prior attempts and constraints in a repository?

The best way to document prior attempts is generating a compact .godmode/research.md file under 60 lines with HIGH, MED, or LOW relevance tags, and appending a .godmode/research-log.tsv entry before planning.

Can I use research to avoid greenfield designs when integrating external libraries?

Yes, you can use this approach to avoid greenfield designs by systematically collecting existing patterns and relevant dependencies. It ensures downstream planning builds on real external context instead of starting from scratch.

Does the research output commit changes before dispatching to think?

Yes, the research output commits the .godmode/research.md file and appends a TSV log entry before dispatching to think. This ensures all prior art and external documentation are saved for downstream planning.