ory/lumen
Cut Claude Code costs and latency with precise, local semantic search across your entire codebase. Best for teams working on large code bases. Enterprise-ready and compliant. No subscription or payment or sign up needed.
Uses Merkle tree change detection and SQLite+sqlite-vec for efficient incremental indexing across 11+ language families via native AST and tree-sitter parsers, splitting code into semantic chunks (functions, types, methods) rather than arbitrary lines. Integrates with Claude Code as an MCP server and supports both Ollama and LM Studio for local embeddings, with git worktree support enabling shared index data across branches that only re-embeds changed files.
Stars
25
Forks
3
Language
Go
License
—
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/mcp/ory/lumen"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
sgroy10/speclock
AI Constraint Engine by Sandeep Roy — stops AI from breaking what you locked. 100/100 on...
patravishek/memex
Retains Claudes session
Intina47/context-sync
Local persistent memory store for LLM applications including continue.dev, cursor, claude...
amanhij/Zikkaron
Biologically-inspired persistent memory engine for Claude Code. 26 cognitive subsystems,...
roampal-ai/roampal-core
Outcome-based persistent memory MCP server for Claude Code and OpenCode. Good advice promoted,...