Memory MCP servers

Read-only local-first MCP server for private Markdown, PDF, and Tika-backed search on Windows.

Persistent memory for AI coding tools with semantic code search, knowledge graphs, impact analysis and decision tracking.

Persistent graph memory that extracts facts from conversations into typed nodes with local embeddings, supporting temporal queries and conflict detection.

Inspectable, correctable project memory in git-versioned Markdown for AI agents.

Embedded local agent memory in one SQLite file per namespace: hybrid retrieval, append-only history queryable at any past time and human-reviewed maintenance.

Memory, docs, and repo context layer for engineering agents.

A lightweight, Git-backed permanent memory for AI agents.

Pay-per-call x402 tools and skill packs (headless browser, web search, OCR, wallet-keyed memory), plus a router that finds the cheapest healthy tool.

Open-source bitemporal memory layer for LLM agents — knowledge graph with conflict resolution.

Local-first memory for coding agents: hybrid LanceDB/Tantivy search, Markdown and code indexes, per-project vaults and recall that abstains on low confidence.

Git-native memory for coding agents: JSON entries in the repo, verified against cited code so hallucinated citations are rejected and stale memory is withheld.

Convert source code repositories into searchable knowledge bases using chunking, OpenAI embeddings and GPT-5 search.

Spec-driven dev workflow MCP: start_* orchestration, memory, GitNexus, quality gates.

Local FAISS vector database for RAG with document ingestion (PDF, TXT, MD, DOCX), semantic search, re-ranking and CLI tools for indexing and querying.

Bhived MCP Server — shared knowledge graph for AI agents. Query, write, and inspect collective AI knowledge through the Model Context Protocol.

Self-evolving memory system for AI agents

Local-first AI memory engine — 4-tier memory, people graph, Bayesian beliefs. Encrypted, 62µs.

Local-first markdown memory system for humans and AI agents: MCP server, typed knowledge graph, deterministic garden planning.

Persistent knowledge graph and MCP server for AI workflows with git-tracked project context.

Temporal memory for AI with decay and reinforcement. Two-layer storage (JSONL + Markdown).

Persistent knowledge memory for AI agents. Hybrid search, code graph, pgvector.

Long-term agent memory backed by a local Obsidian vault: hybrid retrieval with reranking, agentic RAG, wikilink community detection, Obsidian Bases and PDF OCR.

Session memory via a local OpenAI-compatible proxy that extracts tasks, decisions and files into a knowledge graph and checkpoints before context overflow.

Claude Code memory plugin that mines past sessions for decisions, conventions and bug fixes, stores them in an Obsidian vault and briefs new sessions.

Search the web, generate AI responses, and create embeddings with real-time information

Ingest, manage, and retrieve documents for RAG-powered AI applications

Open-source causal memory for AI agents: persistent, explainable, MCP-native memory (13 tools).

Code intelligence for repository graphs, change impact, test selection, and engineering memory.

Persistent memory MCP server. 28 tools, BM25+MMR+OOD retrieval, tamper-evident audit, code graph.

Access Vectara's RAG-as-a-service platform.

Hierarchical agent memory engine with automatic decay, promotion, semantic deduplication and a self-organizing topic tree.

Sync public web sources into cited context packs for AI agents, RAG, and MCP clients.

Pre-action gates that prevent coding agents from repeating known mistakes by turning captured feedback and failures into rules enforced via hooks.

Pre-action gates that block AI agents from repeating known mistakes.

Hosted memory that learns and forgets: feedback re-ranks useful memories, recall fades by recency and similar memories consolidate.

MCP memory server for ZenBrain: agent memory for any Model Context Protocol client — store, recall, consolidate, health. Local SQLite file, no account.

Local-first biological memory for coding agents: recall, consolidation, and pre-action Guard checks.

Local-first memory — 100+ tools, 99.2% LongMemEval-S retrieval@8, hybrid search, GDPR, no cloud.

Security testing tools for penetration testing, network forensics, memory analysis and vulnerability assessment.

Standalone Obsidian vault server: hybrid search, notes and files, memory and tasks; run locally or self-host for remote access.

Persistent semantic and graph-based agent memory in a single self-contained binary with embedded database and models.

RAG Search over your content powered by Inkeep

Local-first memory with verbatim recall, querying, health checks and context-sensitive orientation, using LLM-free retrieval by default.

Semantic knowledge graph for Obsidian with graph, classification and hierarchy modes, local embeddings, semantic bridge discovery and drift detection.

Local repo intelligence: hybrid BM25/vector search, a tree-sitter code graph, queryable git and GitHub history, and persistent source-anchored memories.

Code generator for Arbitrum: Stylus smart contracts, SDK bridging, full-stack dApps and Orbit chain deployment, backed by RAG over Arbitrum documentation.

Private AI context layer for semantic code search, dependency graphs, memory, and workspaces.

Four-tier memory (episodic, semantic, project, procedural) with temporal scoring, contradiction detection, entity tracking and a desktop visualization orb.