Stacklit
One command gives AI agents instant codebase context. ~250 tokens replaces 50,000+ tokens of exploration. Auto-configures Claude Code, Cursor, and Aider.
Every AI coding session starts the same way: the agent burns thousands of tokens just figuring out what your codebase even looks like. Stacklit scans the repo once and generates a compact, AI-agent-ready index -- around 250 tokens replacing 50,000+ tokens of exploration, over 200x smaller.
It uses tree-sitter to parse 11 languages, generates visual dependency maps, and ships an MCP server for AI coding assistants. Merkle hashing enables smart caching so re-scans are instant, and it auto-configures Claude Code, Cursor, and Aider so there's no manual setup step.
Crossed 100 GitHub stars within weeks of release.
Related Projects
AskVerdict AI
AI-powered verdict engine with multi-agent debate that cross-examines every question so you can decide with confidence.
theAuth
Open-source auth for AI agents and humans. Agent identity with scoped permissions and delegation chains, an MCP OAuth 2.1 server, passkeys, SSO and SCIM. TypeScript and Go SDKs.
Claude Code Marketplace
Community-driven marketplace of specialized Claude Code agents covering the full SDLC -- code review, security scanning, deployment automation, and more.