Side-by-side comparison of two Model Context Protocol servers — install paths, tools, quality signals, and directory engagement so you can pick the right one for Claude, Cursor, and other MCP clients.
Compressed Knowledge Graphs (pre-structured dependency DAGs) as MCP context — agents traverse declared edges instead of inferring from text. Open CKG Benchmark (64 domains, 10,838 queries): 3.8× RAG's F1 at 11× fewer tokens, 42× RDS, 0 fabricated edges by construction. 97 domains, 4 tools, no DB/embeddings. pip install ckg-mcp
Self-improving learning engine for Claude Code. Extracts signals from every session (corrections, successes, confusion), runs periodic reflections, and evolves behavioral frameworks through evidence tiers (hypothesis → observed → validated). Frameworks that keep working get promoted, bad ones get retired. 9 MCP tools, automatic hooks, phase-adaptive learning. Single dependency, local-only. npx claude-soul init --starter