memory MCP servers

Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.

Local-first, cloud-friendly memory — 100+ tools, 99.2% LongMemEval-S retrieval@8, hybrid search.

deja-vu: local memory over the session histories of forty-two coding agents.

Agent memory: working, episodic, semantic, procedural and core layers. Local SQLite, no account.

On-device memory layer for AI agents. Claude Code, OpenClaw, and Hermes. Hooks + MCP server + hybrid RAG search.

Session memory for coding agents: local-first recall, drift detection, on-device inference.

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

Local-first persistent session memory for Claude Code and other agents. SQLite + FTS5, no cloud.

Open, local-first engram memory for AI agents: read, correct, delete; shared over MCP.

The private, owned conversation-memory layer for AI. Record, transcribe, and search every meeting.

Local-first knowledge engine for AI agents: memory, graph, and recall over MCP.

Audited memory over MCP: an append-only fact ledger plus verifiable invariants I1-I5.

Local-first shared memory for multiple AI agents that compounds with use; Markdown + git, MCP + CLI.

Cognitive memory for coding agents: hybrid recall, remember/forget, knowledge graph, nightly dreams.

Memory for AI agents: a data bank of MTSV relations, read through one tool.

Git-friendly memory for AI coding: a tree of thread summaries with inherited context

Offline read-only search over a checksum-verified, optionally signed memory pack, with exact quotes

Design transformer LLM architectures and report their parameters, FLOPs, memory and cost

Local-first AI PKM memory for coding conversations.

Heartwood Memory — governed AI agent memory; signed provenance. BUSL-1.1; not OSI open source.

Pay-per-call AI query and versioned task memory for agents. No account, no API key, via x402.

Local personal memory for agents as MCP tools — remember, recall, search, context, graph traversal.