🛠️ kit documentation
Welcome to kit – the Python toolkit from Cased for building LLM-powered developer tools and workflows.
kit shines for getting precise, accurate, and relevant context to LLMs. Use kit to build code reviewers, code generators and graphs, even full-fledged coding assistants: all enriched with the right code context.
MIT-licensed on GitHub.
Map any codebase
Get a structured view with file trees, language-aware symbol extraction (powered by tree-sitter), and dependency insights. Built-in support for 12+ languages, with intelligent caching.
Support for multiple search methods
Mix and match to optimize for speed, accuracy, and use case. Combine fast text search with semantic vector search to find relevant code snippets instantly.
Fine-grained docstring context
Use generated docstrings to find code snippets, answer questions, and improve code generation based on summarized content.
Build AI Workflows
Leverage ready-made utilities for code chunking, context retrieval, and interacting with LLMs. Enhanced with PR summarization and intelligent commit message generation.
Explore the docs
Enhanced MCP server for Cursor, Windsurf, Claude Code & VS Code with deep documentation research and smart context building.
Use kit directly in Claude Code with the official plugin. Claude autonomously uses kit for codebase analysis.
Production-ready pull request reviewer with customizable profiles, CI/CD integration, and context-aware feedback.
Build your own AI developer tools with Kit’s powerful Python library for code intelligence and context building.