Skip to main content

Agent-agnostic memory layer — one memory store for every AI agent.

Project description

AgnoMem

Agent-agnostic memory layer — one memory store for every AI agent.

Claude forgets when you create a new account. GPT-4 has no idea what Claude learned. Cursor and your scripts live in completely separate memory silos. AgnoMem fixes this: a single local memory server that every AI agent reads from and writes to automatically.

pip install agnomem
export OPENAI_API_KEY=sk-...   # or ANTHROPIC_API_KEY
agno install

That's it. Claude Code, Cursor, Windsurf, and your own Python scripts now share the same memory.


How it works

Your AI Agents (Claude, GPT-4, Cursor, custom scripts)
            │
            ▼
     AgnoMem server (localhost:7832)
            │
            ▼
     Mem0 + ChromaDB  (persisted at ~/.agnomem)
  • agno install — installs Mem0, starts the server, injects instructions into Claude Code's CLAUDE.md and your editor's rules file
  • SDK patch — two lines in your Python script and every anthropic / openai call gets relevant memories prepended to the system prompt automatically
  • Portable — all data lives in ~/.agnomem. Export, backup, or move it anywhere.

CLI

agno install                      # full setup in one command
agno start / stop / status        # server management
agno search "how do I deploy?"    # semantic search over memories
agno remember "use fly deploy"    # write a memory
agno list                         # list all memories
agno inject --project             # inject into current project's editor rules
agno eject                        # remove all injections cleanly

Python SDK

import agnomem
agnomem.enable()   # patches Anthropic + OpenAI SDKs — call once at startup

import anthropic
client = anthropic.Anthropic()
# every subsequent call automatically gets relevant memories in the system prompt

Requirements

  • Python 3.10+
  • OPENAI_API_KEY or ANTHROPIC_API_KEY (Mem0 uses an LLM to extract facts)

License

MIT

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

agnomem-0.2.1.tar.gz (18.7 kB view details)

Uploaded Source

Built Distribution

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

agnomem-0.2.1-py3-none-any.whl (21.5 kB view details)

Uploaded Python 3

File details

Details for the file agnomem-0.2.1.tar.gz.

File metadata

  • Download URL: agnomem-0.2.1.tar.gz
  • Upload date:
  • Size: 18.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.9

File hashes

Hashes for agnomem-0.2.1.tar.gz
Algorithm Hash digest
SHA256 22b2591668f6751ea9c8154ff1ff0d12d4b3f461a77c90c96450997f6fd65dc0
MD5 2a6f788a06ccbb17fd5345fbfa179346
BLAKE2b-256 bdcd9375cebe5d19ccf2bf8c2e0dd394e1c6a7e69da9a7e843e462767998259e

See more details on using hashes here.

File details

Details for the file agnomem-0.2.1-py3-none-any.whl.

File metadata

  • Download URL: agnomem-0.2.1-py3-none-any.whl
  • Upload date:
  • Size: 21.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.9

File hashes

Hashes for agnomem-0.2.1-py3-none-any.whl
Algorithm Hash digest
SHA256 04e0c763c38fb2e841461b34a7dc1e47def6c735647be6ab9cc0f8b56638d30d
MD5 17f1006e86797609fe28c17056fa6a10
BLAKE2b-256 3d13001fc352eed302e0699753dae6bab91093cbfb4a62b86bf66bd530446d3e

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