One command gives AI agents instant codebase context, replacing 50,000+ tokens of exploration with ~250 tokens.
Stacklit is easy to set up with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view glincker/stacklit --webOpen the official repository or website.
Check the README for package manager, auth, and platform requirements.
Try it in a small test task inside your agent workflow.
Stacklit is a CLI tool that scans your codebase and generates a compact index file (~250 tokens) that AI agents can read instead of scanning hundreds of files. It automatically configures tools like Claude Code, Cursor, and Aider to use this index, saving context window and speeding up development.
Stacklit is a developer tool that creates a lightweight, AI-readable index of your entire codebase with a single command. It analyzes your project structure, dependencies, exports, and git activity, then outputs a `stacklit.json` file (~250 tokens) that replaces the need for AI agents to scan thousands of lines of code. This drastically reduces token usage and context window consumption, allowing AI coding agents to understand the codebase instantly. The tool auto-detects popular AI coding tools (Claude Code, Cursor, Aider) and configures them to use the index. It also provides an interactive visual map, a Mermaid dependency diagram, and optional MCP server integration. Stacklit is designed for monorepos and large codebases, with support for multiple languages and frameworks. It can be run as a one-time setup or integrated into CI/CD pipelines via a GitHub Action to keep the index fresh.
Strong trust signals; still review the README and permissions before production use.
Last commit was about 31 days ago.
87 GitHub stars indicate community interest.
6 open issues signal maintenance load.
MIT license detected.
Quickly onboard AI agents to large, unfamiliar codebases without manual context setup.
Reduce token costs and context window usage when using AI coding assistants like Claude Code or Cursor.
Maintain an up-to-date codebase index in CI/CD for consistent AI agent performance across the team.
Visualize codebase structure and dependencies with the interactive map and Mermaid diagram.
Automatically configure multiple AI tools (Claude Code, Cursor, Aider) with a single command.
The tool generates a file (stacklit.json) that summarizes your codebase; ensure it does not expose sensitive information if committed to a public repository.
The MCP server integration may require network access; review its configuration for security implications.
87
Stars
10
Forks
6
Issues
MIT
License
Universal AI context generator that saves thousands of tokens per conversation in Claude Code, Cursor, Copilot, and more.
A fast Rust tool to serialize text files in a repo for LLM consumption.
A memory layer for coding agents that stores and retrieves repo context, conventions, and invariants.
2 security/trust notes recorded.
Setup difficulty is 1/5.