mcpField
D
domain-discover

domain-discover

github.com/delexwDevelopment247 installsUpdated 31w ago
Repository

About domain-discover

`domain-discover` is a skill for analyzing a codebase and producing a domain knowledge Markdown file for future Claude Code sessions. It uses repository structure, configuration, application files, and existing documentation as inputs.

Key features

  1. 01Infer domain name and output directory from arguments
  2. 02Scan repository structure and technology stack
  3. 03Review configuration and key application files
  4. 04Consult existing documentation
  5. 05Create or update a domain Markdown file

Use cases

  1. 01Document a codebase domain for future Claude Code work
  2. 02Create a domain file in a specified output directory
  3. 03Refresh domain notes after reviewing project structure

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/delexw/claude-code-misc --skill domain-discover

After installing, type/domain-discoverin Claude Code to confirm it loaded.

Bundled filesreferences/output-format.mdreferences/rules.md

Details

From the project README

Discover Domain Knowledge

You are tasked with creating a comprehensive domain knowledge file for a codebase that will help future Claude Code instances work effectively with this repository.

Inputs

Raw arguments: $ARGUMENTS

Infer from the arguments:

  • DOMAIN_NAME: the domain name, used as the output filename
  • OUT_DIR: output directory, or . (project root) if not provided

Output Location

  • Creates or updates OUT_DIR/{DOMAIN_NAME}.md
  • Run mkdir -p OUT_DIR before writing to ensure the directory exists.

Objective

Analyze the provided codebase and create or update a {DOMAIN}.md file containing essential information for productive development.

Analysis Approach

  1. First, scan the repository structure to understand the project type and technology stack
  2. Read key configuration files (package.json, requirements.txt, Makefile, etc.) to identify build/test commands
  3. Examine main application files to understand the high-level architecture
  4. Review existing documentation (README.md, docs/) for important context

Execution

Follow references/rules.md for content requirements, guidelines, and safety rules. Use references/output-format.md for the output template and validation criteria.

Comments

Markdown supported