michaelalber/grounded-code-mcp
LLMs are sophisticated pattern matchers, not reasoning engines. This project mitigates known failure modes (hallucination, brittle compositionality, lack of grounding) by anchoring AI coding assistants to vetted, authoritative documentation via RAG.
Stars
—
Forks
—
Language
Python
License
MIT
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/mcp/michaelalber/grounded-code-mcp"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
bgauryy/octocode-mcp
MCP server for semantic code research and context generation on real-time using LLM patterns |...
DeusData/codebase-memory-mcp
MCP server that indexes your codebase into a persistent knowledge graph. 64 languages, sub-ms...
wrale/mcp-server-tree-sitter
MCP Server for Tree-sitter
aimasteracc/tree-sitter-analyzer
A scalable, multi-language code analysis framework based on Tree-sitter, usable both as a CLI...
duysolo/codebaxing
Local codebase indexing and semantic code search via MCP