The definitive benchmark for agent memory systems. 56 tests, 8 categories, scale testing up to 10K memories.

MnemoDB — the agent memory format. A database whose files you can read: human-readable .mem.md memory for AI agents. Umbrella package re-exporting @mnemodb/core.

Associative memory for LLM agents — a key-graph memory layer (MCP server) that recalls by association, not just vector similarity. A graph-based alternative to vector-store agent memory.

TypeScript port of tcvdb_text — BM25 sparse vector encoder for Tencent Cloud VectorDB

Pi package for AgentDir local-first coding-agent memory and evidence capture.

mnemo — CLI for MnemoDB agent memory stores: init, list, show, doctor.

AI agent memory & session orchestrator for MCP — persistent KV-Cache, Soul Board, immutable Ledger

Cross-Agent Memory for Projects — local-first project context for coding agents.

Portable agent memory in a git repo you own — the canonical agent history & memory store (canon): sync, translate, pull, capability artifacts, and project knowledge graphs across AI coding harnesses.

Local memory package with SQLite store, embedding search, and Bullet-Journal engine subpaths.

AI-friendly database with vector search, semantic indexing, and agent memory

The MemorySync CLI. Add, search and manage agent memory from your terminal, with a JSON mode built for AI agents.

Four-layer local memory system plugin for OpenClaw — auto-captures, structures, and profiles conversational knowledge using local LLM + SQLite vector search (L0→L1→L2→L3 pipeline)

Durable agent memory for the Pi coding agent: every completed turn becomes an owner-controlled Markdown episode, with exact, ranked, bounded recall.

Stratus agent identities: one-call creation with generated names and avatar themes, per-agent memory, delegation, and routing

Persistent self-improving agent memory — install & forget: auto-inject context via git hooks, zero-arg bootstrap, machine-wide global vault, 13 memory types, confidence tracking, XML codebase graphs

Durable per-project agent memory and plan tracking in Org mode

Bluud — agent memory for this project. The npx @bluud/cli CLI installs the Bluud skill into your AI tools and keeps project memory in sync.

Spinal Plug client: durable, project-scoped agent memory with local-first storage and selective sync.

Agent memory with no API key, no LLM, and no embedding provider. Serves MCP over stdio against a local SQLite store, or Cloudflare Workers + D1. Drop-in for @modelcontextprotocol/server-memory; every memory keeps its source, its scope, and the evidence th