// codegen.com

Find tools. Build agents.

The developer directory for Claude Code, Cursor, Windsurf, and every AI coding tool worth using.

Browse the directory → Start a course →

codegen — zsh

$ codegen search "summarize a PR"

# 3 results across the directory

→ diff-summarizer skill Claude Code, Cursor

→ code-review-agent agent ~20 min build

→ using-plan-mode guide 9 min read

$ claude skill add diff-summarizer

✓ installed diff-summarizer — ready in this session

CL Claude Code Anthropic terminal first autonomous coding agent with 72.5% SWE bench score, up to 1M token context, and native MCP integration. CU Cursor AI native code editor built on VS Code with Composer for multi file editing, parallel agents, and the largest community in the category. GI GitHub Copilot GitHub AI coding assistant with inline suggestions, chat, and expanding agentic capabilities. The most widely adopted tool with the lowest entry price. WI Windsurf AI native IDE with Cascade agent mode and adaptive Memories that learn your coding style, now owned by Cognition AI. CO Codegen Enterprise orchestration platform for AI coding agents with full ClickUp task context, governance, and audit trails. DE Devin Autonomous AI software engineer that plans, codes, tests, and delivers pull requests in a sandboxed environment.

View all 18 tools → Browse 1 skills →

// comparisons

Claude Code vs Windsurf: Which AI Coding Agent Fits Your Workflow? Claude Code and Windsurf both run AI agents that read your codebase, plan changes, and edit across many files. GitHub Copilot vs Windsurf: Which AI Coding Tool Should You Pick in 2026? Copilot and Windsurf represent two fundamentally different approaches to AI-assisted coding. Antigravity CLI vs Claude Code: Which Terminal Coding Agent Wins in 2026? Both tools live in the terminal and drive an agent that reads your code, edits files, and runs commands.

All comparisons →

// learn

code-review.yml

name: code-review
on: pull_request
skills:
  - diff-summarizer
  - security-lint
  - test-coverage
rules:
  - review-checklist
  - no-secrets
model: claude-sonnet-4.5
comment: on-failure
run: claude review --diff origin/main

// agent recipes

Build automated workflows

01Reads the PR diff

02Summarizes changes across files

03Runs security and secrets scan

04Checks test coverage

05Posts a review comment on failure

Browse agent recipes →

Pick a tool

Every tool page has install commands, config files, and guides ready to copy.

Browse the directory →