skills.rest
Visma-Tech-LT
Visma-Tech-LTOFFICIAL
Explore all skills by this author →
GitHub

research-codebase

Research unfamiliar codebases and produce a self-contained markdown artifact with evidence.

3|Updated Apr 10, 2026
Search Tags:#implementation planning#parallel agents#technical documentation#architecture analysis#codebase research#evidence-based findings#repo discovery
One-click install
npx skills add https://github.com/Visma-Tech-LT/ai-workshop-marketplace --skill research-codebase-visma-tech-lt
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research-codebase
Source: https://github.com/Visma-Tech-LT/ai-workshop-marketplace/tree/main/plugins/sdw/skills/research-codebase
Command: npx skills add https://github.com/Visma-Tech-LT/ai-workshop-marketplace --skill research-codebase-visma-tech-lt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you quickly understand an unfamiliar or complex repository so you can answer questions and make implementation decisions with confidence instead of guessing.

Core Features & Use Cases

  • Comprehensive codebase research: Decomposes a question into focused research areas and gathers evidence with concrete file path references.
  • Parallel sub-agent discovery: Locates relevant components first, then analyzes the most promising findings to uncover implementation details and architecture.
  • Self-contained research artifacts: Produces a complete research-YYYY-MM-DD-description.md document intended to be used as the contract for later planning and implementation.

Quick Start

Ask to research a specific topic by providing your research question, and then review the generated research document that will be saved under .claude/ai/.

Frequently Asked Questions about research-codebase

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

FAQPage Schema
How do I research an unfamiliar codebase to understand its architecture?▼

To research an unfamiliar codebase, decompose your technical question into focused areas and use parallel agents to locate relevant components, gathering evidence with concrete file path references to synthesize architecture findings.

What's the best way to plan implementation across a repository with multiple components and dependencies?▼

Implementation planning across a complex repository is best done by generating a research markdown artifact with decomposed findings, cross-references, and resolved open questions to serve as the contract for later implementation steps.

How do I prepare evidence-based technical documentation for an unknown repository?▼

Prepare evidence-based technical documentation by analyzing the most promising findings from parallel sub-agent discovery, resolving open questions, and producing a self-contained research markdown document with cross-references.

Can I use parallel agents for repo discovery and architecture analysis on any sizable codebase?▼

Yes, parallel agents can be used for repo discovery and architecture analysis across any sizable repository with multiple components and dependencies, locating relevant components first before analyzing implementation details.

Where are the generated codebase research artifacts saved?▼

Generated codebase research artifacts are saved as a complete research-YYYY-MM-DD-description.md document under the .claude/ai/ directory, intended to be used as the contract for later planning and implementation.

Related Skills

Explore More Software Engineering Skills→
dformosodformoso

to-prd

Generate a Product Requirements Document from conversation, codebase, and research.

Community
Advanced
jineynejineyne

speckit-plan

Generate implementation plans and design artifacts from spec-kit repositories.

Community
Advanced
jungmini0601jungmini0601

planning

Plan technical implementations and design system architectures with roadmaps.

Community
Advanced
visualityplvisualitypl
4

research_codebase_nt

Document existing codebase architecture and functionality into structured markdown.

Official
Advanced