Skip to main content

Local memory server for AI agents — offline, private, fast

Project description

memra-local

Local-first memory server for AI agents. Offline, private, fast.

memra-local gives your coding agent persistent memory that lives entirely on your machine — no account, no network, no data leaves the laptop. Works with Claude Code, Cursor, Zed, Droid, Hermes Agent, OpenClaw, and any MCP-compatible client.

When you're ready to sync across devices or share with a team, a single command pushes your local namespace to Memra Cloud. Same tools, same API, your choice.

Install

pip install memra-local
memra mcp          # start the MCP server

Requires Python 3.10+.

Wire it into your editor

Claude Code / Cursor

{
  "mcpServers": {
    "memra": {
      "command": "memra",
      "args": ["mcp"]
    }
  }
}

Zed

{
  "context_servers": {
    "memra": {
      "command": { "path": "memra", "args": ["mcp"] }
    }
  }
}

Droid (Factory.ai) / Hermes Agent / OpenClaw

See usememra.com/install for client-specific snippets.

What you get

  • Flat-file memory in ~/.memra/ — plain YAML, inspectable, greppable, diff-able
  • MCP server exposing memra_add, memra_recall, memra_get, memra_list, memra_supersede, memra_history, and more
  • Local embeddings via fastembed (ONNX multilingual-e5-small, ~100 languages) — no OpenAI key, no PyTorch
  • Sync to cloud optional: memra sync enable <namespace> --api-key memra_live_...

Commands

memra mcp          # MCP server over stdio
memra status       # store health: scope, memory count, disk usage, sync
memra hooks install  # optional — auto-capture decisions/patterns as you work
memra --help       # full CLI reference

Docs + source

License

BUSL-1.1. Change Date 2030-04-17 — on that date the license auto-converts to Apache-2.0. Until then, personal and non-production use are unrestricted; commercial production use requires a separate license.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

memra_local-4.6.1.tar.gz (46.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

memra_local-4.6.1-py3-none-any.whl (53.3 kB view details)

Uploaded Python 3

File details

Details for the file memra_local-4.6.1.tar.gz.

File metadata

  • Download URL: memra_local-4.6.1.tar.gz
  • Upload date:
  • Size: 46.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.3

File hashes

Hashes for memra_local-4.6.1.tar.gz
Algorithm Hash digest
SHA256 8fd58155d2cbf0de380058fe21d1d8f2be6dab06faf5694e90c56d02ac01af14
MD5 b13e023d99a95e21eb98939f3d0249a3
BLAKE2b-256 ef6c9f13bd096eb2756a2b25738849a10507d2f6937f5487d662b32d08b5fb17

See more details on using hashes here.

File details

Details for the file memra_local-4.6.1-py3-none-any.whl.

File metadata

  • Download URL: memra_local-4.6.1-py3-none-any.whl
  • Upload date:
  • Size: 53.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.3

File hashes

Hashes for memra_local-4.6.1-py3-none-any.whl
Algorithm Hash digest
SHA256 fd15c60d2b2458d27013a34d439b642c055b777c3ccb346df9d9f723c61b5761
MD5 d06b714c57bf55fe88b61c1a92103a0c
BLAKE2b-256 0e08ff2da5841650e76e9c9f511b2606d6ae1fb93f2c0c61c8bfbc2196efa531

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page