project-explain

Explains project concepts by reading code and citing real file locations.

1|Updated Sep 11, 2026
One-click install
npx skills add https://github.com/lldwb/lldwb-claude-skills --skill project-explain-lldwb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-explain
Source: https://github.com/lldwb/lldwb-claude-skills/tree/main/skills/project-explain
Command: npx skills add https://github.com/lldwb/lldwb-claude-skills --skill project-explain-lldwb

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? When you encounter unfamiliar concepts, mechanisms, or flows in a codebase, generic explanations from documentation or tutorials do not tell you how they actually work in your project. This Skill reads your project's code and docs first, then explains each item grounded in real implementations with exact file and line references. ## Core Features & Use Cases - Project-grounded explanations: Each concept is covered in four parts—what it is, how it is implemented in this project, key code locations (file:line), and project-specific pitfalls and conventions. - No fabrication: Conclusions without code evidence are explicitly marked as unverified rather than guessed; line numbers are re-read before delivery. - Structured output: Delivers Markdown with a top-level title, itemized sections, a closing "what is this project" summary, and a reference table of cited locations. - Use Case: You join a legacy Java backend project and ask it to explain the permission-check flow and the caching mechanism; it traces the call chains, cites exact classes and lines, and flags conventions unique to this repo. ## Quick Start Explain how the authentication flow and the retry mechanism work in this project, citing the actual code locations.

Frequently Asked Questions about project-explain

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

FAQPage Schema
How do I understand an unfamiliar codebase quickly?

Ask for explanations of specific concepts or flows rather than a generic overview. The Skill reads the project's README, build files, and relevant code first, then explains each item with exact file and line citations so you can verify every claim.

How to explain code with accurate file and line references?

The Skill locates implementations via keyword search on class names, methods, and config keys, records file:line positions, and re-reads them before delivery to confirm accuracy. Large projects can be split across read-only subagents per module.

Can it modify code or write documentation files?

No, it only produces explanation content in the conversation and never edits code or creates files. Persisting explanations as project documentation is delegated to a separate doc-sync skill, and code changes go to bug-fix or feature-dev skills.

What happens when something cannot be found in the code?

Unverifiable points are explicitly marked as not found in the code rather than guessed or omitted. Speculative conclusions are labeled as unconfirmed with the supporting evidence and a suggested verification method.

Does it work for large multi-module projects?

Yes. For content spanning many modules, it can dispatch read-only exploration subagents per module to collect facts with file:line references, then the main agent consolidates them into one structured explanation.