Home > RAG & Memory > memory-engine

memory-engine

Agentic memory built on Postgres

agentic-memory bm25 rag semantic-search typescript

Why this rank:Recent releaseHealthy release cadenceStrong adoption

Description

Agentic memory built on Postgres

README

Memory Engine

Permanent memory for AI agents. Store, search, and organize knowledge across conversations.

Memory Engine gives AI coding agents a persistent memory layer they can read from and write to via MCP. Memories are organized with tree paths, tagged with metadata, and searchable by meaning (semantic), keywords (BM25), or both (hybrid via Reciprocal Rank Fusion).

Install

curl -fsSL https://install.memory.build | sh

Alternative install methods:

Homebrew

brew install timescale/tap/me

NPM

npm i -g @memory.build/cli

Quick start

# Authenticate
me login

# Store a memory
me memory create "Auth uses bcrypt with cost 12" --tree design.auth

# Search by meaning
me memory search "how does authentication work"

# Connect to your AI tools (Claude Code, Gemini, Codex, OpenCode)
me mcp install

How it works

Memory Engine runs as an MCP server that AI agents connect to over stdio. Each agent gets 10 tools for creating, searching, and managing memories. All data lives in PostgreSQL, using native extensions for search:

pgvector for semantic (vector) search
pg_textsearch for BM25 keyword search
ltree for hierarchical tree paths
JSONB + GIN for metadata filtering
tstzrange for temporal queries
Row-Level Security for access control

Documentation

Getting Started -- install, login, first memory
Core Concepts -- memories, tree paths, metadata, search modes
Access Control -- users, roles, grants, ownership
Memory Packs -- pre-built knowledge collections
MCP Integration -- connecting AI agents

Reference

CLI Commands -- full command reference
MCP Tools -- full MCP tool reference

License

Apache 2.0

Release History

Version	Changes	Urgency	Date
v0.8.0	## What's Changed * docs: add living system design by @jgpruitt in https://github.com/timescale/memory-engine/pull/210 * build(postgres): add TimescaleDB by @jgpruitt in https://github.com/timescale/memory-engine/pull/211 * feat(memory): append-only audit log with history reads, revert, and undelete by @jgpruitt in https://github.com/timescale/memory-engine/pull/212 * docs: event-log design doc, design/ in CLAUDE.md, and 0.8.0 changelog by @jgpruitt in https://github.com/timescale/memory-engine/	High	8/13/2026
v0.7.0	## What's Changed * Add API key TTL flag by @jgpruitt in https://github.com/timescale/memory-engine/pull/170 * Track API key last-used dates by @jgpruitt in https://github.com/timescale/memory-engine/pull/171 * feat(auth): add scoped API keys for least-privilege automation by @jgpruitt in https://github.com/timescale/memory-engine/pull/172 * chore(deps): bump better-auth from 1.6.20 to 1.6.22 in /packages/web by @dependabot[bot] in https://github.com/timescale/memory-engine/pull/173 * fix(server	High	8/5/2026
v0.6.2	## What's Changed * fix(server): serve invite links as SPA routes by @jgpruitt in https://github.com/timescale/memory-engine/pull/165 * docs: add project setup guide by @jgpruitt in https://github.com/timescale/memory-engine/pull/166 * docs: true up harness integration docs by @jgpruitt in https://github.com/timescale/memory-engine/pull/167 * feat(web): dark mode + favicon by @nateships in https://github.com/timescale/memory-engine/pull/168 * feat(space): member-accessible space.listMembers + `m	High	7/16/2026
v0.4.0	## What's Changed * fix(cli): flatten newlines in `me search` table output (TNT-153) by @jgpruitt in https://github.com/timescale/memory-engine/pull/110 * fix(cli): render memory tree root as `/` instead of `.` by @jgpruitt in https://github.com/timescale/memory-engine/pull/111 * Redesign web UI: Tiger Data "Console" (light) by @cevian in https://github.com/timescale/memory-engine/pull/109 * fix(worker): stop amplifying embedding-provider rate limits by @jgpruitt in https://github.com/timescale/	High	7/6/2026
v0.3.3	## What's Changed * fix(release): repair protocol build entry list + gate cli publish on publish-npm by @cevian in https://github.com/timescale/memory-engine/pull/105 Full Changelog: https://github.com/timescale/memory-engine/compare/v0.3.2...v0.3.3	High	6/25/2026
v0.2.6	## What's Changed * Dedup planned memories by id before batchCreate by @graveland in https://github.com/timescale/memory-engine/pull/63 * Skip duplicate ids in batchCreateMemories instead of aborting by @graveland in https://github.com/timescale/memory-engine/pull/64 * Chunk batchCreate by byte budget, not just count by @graveland in https://github.com/timescale/memory-engine/pull/65 * Make request body size limit configurable via env by @graveland in https://github.com/timescale/memory-engine/p	High	5/5/2026
v0.2.5	## What's Changed * Add `me apikey show` to print stored key from credentials.yaml by @jgpruitt in https://github.com/timescale/memory-engine/pull/61 Full Changelog: https://github.com/timescale/memory-engine/compare/v0.2.4...v0.2.5	High	4/29/2026
v0.2.1	## What's Changed * add client/server version compatibility check by @jgpruitt in https://github.com/timescale/memory-engine/pull/49 * fix delete/move tree dry-run count cap at 1000 (TNT-59) by @jgpruitt in https://github.com/timescale/memory-engine/pull/50 * replace mkdocs with custom next.js docs site by @jgpruitt in https://github.com/timescale/memory-engine/pull/52 * Add semantic search threshold by @murrayju in https://github.com/timescale/memory-engine/pull/51 * Use hybrid search for CLI p	High	4/27/2026
v0.1.17	## What's Changed * Me0/mcp optional by @jgpruitt in https://github.com/timescale/memory-engine/pull/41 * decouple client and server releases by @jgpruitt in https://github.com/timescale/memory-engine/pull/42 Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.16...v0.1.17	High	4/22/2026
v0.1.16	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.15...v0.1.16	High	4/20/2026
v0.1.15	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.14...v0.1.15	High	4/20/2026
v0.1.14	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.13...v0.1.14	High	4/17/2026
v0.1.13	## What's Changed * style docs site to match marketing site branding by @jgpruitt in https://github.com/timescale/memory-engine/pull/39 * add identity attributes to rpc spans, disable biome lint for docs sty… by @jgpruitt in https://github.com/timescale/memory-engine/pull/40 Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.12...v0.1.13	High	4/17/2026
v0.1.12	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.11...v0.1.12	High	4/16/2026
v0.1.10	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.9...v0.1.10	High	4/16/2026
v0.1.9	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.8...v0.1.9	High	4/16/2026
v0.1.8	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.7...v0.1.8	High	4/16/2026
v0.1.7	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.6...v0.1.7	High	4/16/2026
v0.1.6	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.5...v0.1.6	High	4/16/2026
v0.1.5	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.4...v0.1.5	High	4/16/2026
v0.1.4	## What's Changed * Me0/docs by @jgpruitt in https://github.com/timescale/memory-engine/pull/38 Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.3...v0.1.4	High	4/16/2026
v0.1.3	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.2...v0.1.3	High	4/15/2026
v0.1.2	Full Changelog: https://github.com/timescale/memory-engine/compare/v0.1.1...v0.1.2	High	4/14/2026
v0.1.1	## What's Changed * feat(engine): move embedding queue from shared schema to per-engine by @jgpruitt in https://github.com/timescale/memory-engine/pull/1 * refactor(embedding): simplify to OpenAI + Ollama with O(1) character-based truncation by @jgpruitt in https://github.com/timescale/memory-engine/pull/2 * add deploy process by @murrayju in https://github.com/timescale/memory-engine/pull/3 * feat: add accounts migration system and enhance engine migrations by @jgpruitt in https://github.com/ti	Medium	4/14/2026
v0.1.0	## What's Changed * feat(engine): move embedding queue from shared schema to per-engine by @jgpruitt in https://github.com/timescale/memory-engine/pull/1 * refactor(embedding): simplify to OpenAI + Ollama with O(1) character-based truncation by @jgpruitt in https://github.com/timescale/memory-engine/pull/2 * add deploy process by @murrayju in https://github.com/timescale/memory-engine/pull/3 * feat: add accounts migration system and enhance engine migrations by @jgpruitt in https://github.com/ti	Medium	4/14/2026

Dependencies & License Audit

Loading dependencies...

Similar Packages

antflyNo descriptionv0.2.1

engineering-notebookCapture and summarize Claude Code sessions into searchable, browsable engineering journals with a web UI and automated daily entries.main@2026-09-07

taleThe Sovereign AI Platform — Local AI models, agents, skills, and automations — on your own infrastructure, connected to your datav0.4.18

mossOfficial Repo of Mossv0.6.2

showcaseShowcase delivers a modern developer portfolio built with TypeScript and React, focusing on interactivity and clean architecture for a seamless user experience.main@2026-06-16

More from timescale

pg-aiguideMCP server and Claude plugin for Postgres skills and documentation. Helps AI coding tools generate better PostgreSQL code.

More in RAG & Memory

pytorch-lightningPyTorch Lightning is the lightweight PyTorch wrapper for ML researchers. Scale your models. Write less boilerplate.

banksA prompt programming language

opikDebug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.

pdf_oxideThe fastest PDF library for Python and Rust. Text extraction, image extraction, markdown conversion, PDF creation & editing. 0.8ms mean, 5× faster than industry leaders, 100% pass rate on 3,830 PDFs.