wren-engine and Context-Engine
About wren-engine
Canner/wren-engine
🤖 The Semantic Engine for Model Context Protocol(MCP) Clients and AI Agents 🔥
Provides semantic modeling through MDL (Modeling Definition Language) to capture business definitions, metrics, relationships, and governance rules that agents can reason over—moving beyond raw schema discovery. Built as an MCP server with connectors to 15+ data sources (Snowflake, BigQuery, PostgreSQL, DuckDB, etc.) and designed for embedding in agent workflows, Claude Desktop, and developer IDEs. Uses Apache DataFusion for query planning and execution, enabling agents to translate natural language into governed, contextualized data access rather than ad hoc SQL generation.
About Context-Engine
Context-Engine-AI/Context-Engine
Context-Engine MCP - Agentic Context Compression Suite
Provides 30+ MCP tools for semantic code search, symbol graph navigation, memory persistence, and cross-repo tracing—packaged as reusable skills that integrate with Claude, Cursor, Windsurf, Gemini, and other AI assistants. The architecture uses batch query operations for ~75% token savings and combines semantic search, symbol graph queries, git history analysis, and structural pattern matching into a unified `search` tool that auto-routes to the optimal backend. Available as a hosted service with VS Code integration and npm bridge (`ctx-mcp-bridge`), or deployed via MCP server protocol.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work