CLI Reference

Usage

codeilus [PATH] [COMMAND]

If PATH is given without a subcommand, Codeilus runs analyze then serve.

Commands

codeilus analyze <path>

Run the full analysis pipeline on a repository.

codeilus analyze ./my-project

Steps: parse, store, graph, metrics, analyze, diagram, narrate, learn.

codeilus serve

Start the interactive web server.

codeilus serve --port 4174

Flag	Default	Description
--port, -p	4174	Port to listen on

codeilus <path>

Shorthand for analyze + serve:

codeilus ./my-project
# Equivalent to:
# codeilus analyze ./my-project && codeilus serve

codeilus harvest

Scrape GitHub trending repositories.

codeilus harvest --trending --languages rust,python

Flag	Description
--trending	Scrape trending repos
--date	Date to harvest (YYYY-MM-DD)
--languages	Filter by languages (comma-separated)

codeilus export

Export analyzed repos as static HTML.

codeilus export ./my-project --output ./output

Flag	Default	Description
--output, -o	./output	Output directory
--all-harvested		Export all harvested repos
--date		Date for harvested repos

codeilus deploy

Deploy static output to CDN.

codeilus deploy ./output --cloudflare

Flag	Description
--cloudflare	Deploy to Cloudflare Pages
--gh-pages	Deploy to GitHub Pages

codeilus mcp

Start the MCP stdio server for AI agent integration.

codeilus mcp

See MCP Server for available tools.