Local memory for Claude Code. Semantic recall across session history with zero API calls.
37-tool MCP server: context health, memory, reasoning, truthfulness & research pipeline.
Local-first memory for Claude Code and any MCP client: hybrid search + knowledge graph, $0/token.
Verification and memory layer for coding-agent workflows.
FastMCP server for surgical queries against a vault knowledge graph (NetworkX node-link JSON).….
Search 120,000+ recalled products from 8 global safety agencies using AI similarity.
Self-hosted AI memory server. Persistent memory on your hardware, not the cloud.
Cross-agent memory over MCP: hybrid recall, knowledge graph, private/shared visibility, redaction.
Persistent memory layer for AI coding agents — your agent remembers what you taught it.
Six pure, in-memory CSS color tools: parse, convert, WCAG/APCA contrast, gamut map, ramps, solver.
Persistent memory, task coordination, and HITL queue for AI coding sessions.
Long-term memory for AI agents: semantic facts, episodic events, and procedural workflows.
Coordination memory with verification: reconcile-against-GitHub, verdict freshness, provenance.
Persistent memory for any AI — zero token cost until recall.
Persistent agent memory as a single Go binary: knowledge graph, TUI observability, zero infra.
Persistent episodic memory for AI coding: capture, retrieve, brief, dream cycle, cross-project.
Live Linux host signal (CPU, memory, processes, network, optional eBPF) for MCP clients.
Private AI context layer for semantic code search, dependency graphs, memory, and workspaces.
AI-powered localization — translate, manage translation memory, and access style guides.
Cross-project knowledge graph for MCP — map features across mobile, backend, and admin codebases.
Shared memory store for AI coding tools with dual-timeline conflict arbitration.
Persistent semantic memory API for AI agents with Audio Intelligence.
Local-first, zero-knowledge AI memory across Claude, Cursor, and every MCP client. 0 RTT reads.
Portable, user-owned AI memory packs: read, validate, fingerprint, diff, and observe.
About Memory MCP servers
Memory servers solve the blank-slate problem: every new conversation starts without the context of the last one. They store facts, preferences and documents in a form the agent can recall, from simple local note stores to knowledge graphs and vector databases such as Chroma, Qdrant, Pinecone and Weaviate. Some index your own files; others manage a shared memory across tools and sessions.
Decide where memories live before you decide which server to use. Local servers keep everything on your machine, which is ideal for personal and sensitive context; hosted servers make memory available to every client you use. Look for servers that let you inspect, edit and delete what has been stored, and that separate retrieval tools from write tools.