"Hybrid Memory Models Combining Relational, Graph, and RAG Approaches" matching MCP servers:

Classifies development task complexity (LIGHT/MEDIUM/HEAVY) and recommends the most cost-efficient AI model per provider, enabling optimized model selection for coding tasks.

MCP server that lists available GitHub Models and compares responses from multiple models for the same prompt.

An MCP server that helps interact with and compare different GitHub Models, enabling listing, filtering, and comparing model responses.

Graph-based tool retrieval for LLM agents. Builds a tool graph from OpenAPI/MCP specs and retrieves multi-step workflows via hybrid search (BM25 + graph traversal + embedding), recovering accuracy from 12% to 82% with 79% fewer tokens. Also works as an MCP Proxy to aggregate multiple servers behind 3 meta-tools.

Enables a model to call the Microsoft Graph API directly through nine verb-shaped tools with per-user on-behalf-of authentication, so users can read, create, update, delete, search, batch, upload, download, and resolve attachments without server-side business logic or allow-lists.

Enables any MCP-capable LLM client to search self-hosted long-term memory over markdown and PDF documents, combining dense semantic vectors with BM25 keyword retrieval and optional cross-encoder reranking. Exposes a read-only tool surface for querying incidents, runbooks, and other knowledge-base content, while writes happen out-of-band through ingestion jobs or a token-gated internal API.

Enables deterministic task routing, contract enforcement, and effect authorization across domain owners while exposing a minimal MCP interface and read-only observability.

MCP server that integrates Apache Ranger authorization with RAG, enforcing Ranger policies to control access to knowledge bases before forwarding queries to RAG Studio.

MCP server for querying the SEC Ownership Graph, providing curated read-only tools to analyze activist campaigns, control chains, and board interlocks from SEC filings. Answers cite accession numbers and abstain when data is insufficient.

MCP server for Parks on the Air (POTA) providing live activator spots, park details, activator/hunter stats, and scheduled activations through any MCP-compatible AI assistant.

MCP server that exposes unified AI-friendly tools over Messari's standardized lending subgraphs on The Graph. One natural-language query → fan out across 40+ lending protocols on multiple chains → get back structured, comparable data.

MCP server for querying Lido liquid staking protocol data via The Graph's decentralized network, providing tools for staking stats, APR, withdrawals, node operators, and governance.

Enables coding agents to retrieve project context from an AnythingLLM RAG at session start and write back human-approved durable facts via five workflow-shaped tools.

Provides a unified MCP interface over Uniswap V2, V3, and V4 across multiple chains via The Graph, enabling agents to query token prices, top pools, pairs, and recent swaps with clean JSON output.

Helps construct and validate Microsoft Graph REST API calls with on-demand tool loading, providing request details, required permissions, and code examples for categories like users, files, mail, calendar, groups, notes, tasks, sites, and subscriptions.

MCP server for querying AAVE V2/V3 lending protocol and governance data via The Graph subgraphs. Exposes 14 tools and 5 guided prompts that any AI agent (Claude, Cursor, Copilot, etc.) can use to query lending markets, user positions, health factors, liquidations, flash loans, rate history, and AAVE governance — across 7 chains (Ethereum, Base, Arbitrum, Polygon, Optimism, Avalanche, Fantom) via

An MCP server that enables querying Polymarket prediction market data through The Graph subgraphs. It provides tools for accessing market stats, trader P\&L, user positions, and orderbook activity using AI agents.

Persistent project memory for AI models and coding agents. Memory MCP stores architecture, decisions, tasks, warnings, preferences, and session state in Supabase so OpenCode, Claude Code CLI, Qwen Code, Codex, or any MCP-compatible client can resume work without losing context.

AI memory server with 53 tools. Knowledge Graph, semantic search, session tracking, multi-agent support.

A portable, user-owned memory for MCP-capable agents in VS Code, enabling cross-agent recall and memory creation.