code-graph-context and deep-code-reasoning-mcp
These are complements: one builds indexable code graphs for context injection while the other provides AI-powered reasoning over code, so they could be chained together where the graph context feeds into the reasoning engine.
About code-graph-context
andrew-hernandez-paragon/code-graph-context
A Model Context Protocol (MCP) server that builds rich code graphs to provide deep contextual understanding of TypeScript codebases to Large Language Models.
Builds Abstract Syntax Trees (AST) using ts-morph, stores relationships in Neo4j, and generates vector embeddings locally (or via OpenAI) to enable semantic search and natural language querying. Includes a dual-schema system with both AST-level and framework-specific nodes (e.g., NestController, HttpEndpoint), alongside features like incremental parsing, workspace auto-detection, impact analysis, and dead code detection.
About deep-code-reasoning-mcp
haasonsaas/deep-code-reasoning-mcp
A Model Context Protocol (MCP) server that provides advanced code analysis and reasoning capabilities powered by Google's Gemini AI
Implements an intelligent multi-model escalation strategy where Claude Code handles local refactoring while Gemini's 1M-token context window analyzes large-scale distributed system failures, logs, and traces that exceed Claude's capacity. Features AI-to-AI conversational analysis tools enabling iterative problem-solving between models, plus specialized tools for execution tracing, cross-system impact modeling, and performance bottleneck detection. Integrates directly with Claude Desktop via MCP protocol and Google's Gemini 2.5 Pro API for complementary code reasoning workflows.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work