Skip to main content

MCP-native memory database for AI agents

Project description

mnemo-db

PyPI distribution of Mnemo — an MCP-native memory database for AI agents.

pip install mnemo-db

The package directory and import path are still mnemo — only the PyPI distribution name has the -db suffix because the unqualified mnemo is held by an unrelated 2021 notebook project. Your code keeps saying:

from mnemo import MnemoClient

Quick start

from mnemo import MnemoClient

client = MnemoClient(db_path="agent.mnemo.db", agent_id="agent-1")

# Three primitives carry most workflows.
result = client.remember("The user prefers dark mode", tags=["preference"])
memories = client.recall("user preferences", limit=5)
client.forget([result["id"]])

# Mem0-compatible aliases also work:
# client.add(...), client.search(...), client.delete(...)

What's in the box

Surface What
MnemoClient Native PyO3 binding to the Rust engine — DuckDB storage, USearch HNSW vector index, Tantivy full-text index, hybrid retrieval, hash-chained audit log
MnemoMemoryToolServer Anthropic memory_20250818 6-op handler. pip install 'mnemo-db[anthropic-memory-tool]'
MnemoLettaShared Letta-style Conversations adapter for shared agent memory
S3Workspace / CloudflareR2Workspace OpenAI Agents SDK GA snapshot store backends. pip install 'mnemo-db[openai-sandbox-s3]' or [openai-sandbox-r2]
MnemoAgentMemory (OpenAI), Mem0Compat, ASMDCheckpointer (LangGraph), 12 more Drop-in framework integrations. Install the matching extra.

Optional extras

pip install 'mnemo-db[langgraph]'              # LangGraph checkpoint
pip install 'mnemo-db[crewai]'                 # CrewAI memory
pip install 'mnemo-db[openai-agents]'          # OpenAI Agents SDK
pip install 'mnemo-db[claude]'                 # Claude Agent SDK
pip install 'mnemo-db[anthropic-memory-tool]'  # memory_20250818
pip install 'mnemo-db[openai-sandbox-s3]'      # S3 workspace backend
pip install 'mnemo-db[openai-sandbox-r2]'      # Cloudflare R2 workspace backend
pip install 'mnemo-db[benchmark]'              # LoCoMo / LongMemEval harness

Each extra fails closed if not installed — from mnemo import MnemoMemoryToolServer will raise ImportError cleanly when [anthropic-memory-tool] isn't on the path.

License + source

Source: https://github.com/sattyamjjain/mnemo. Apache-2.0.

Documentation: https://github.com/sattyamjjain/mnemo#readme. Issues: https://github.com/sattyamjjain/mnemo/issues.

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

mnemo_db-0.4.0rc2.tar.gz (235.8 kB view details)

Uploaded Source

Built Distribution

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

mnemo_db-0.4.0rc2-cp312-cp312-macosx_11_0_arm64.whl (17.6 MB view details)

Uploaded CPython 3.12macOS 11.0+ ARM64

File details

Details for the file mnemo_db-0.4.0rc2.tar.gz.

File metadata

  • Download URL: mnemo_db-0.4.0rc2.tar.gz
  • Upload date:
  • Size: 235.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.7

File hashes

Hashes for mnemo_db-0.4.0rc2.tar.gz
Algorithm Hash digest
SHA256 b4fb9d057b581bc0373bafce20fdc8bbec7fbb41a9560d4caeac6eeba30ec06e
MD5 9fc7623669f1323b1e4a1154b459c4eb
BLAKE2b-256 ee3ed047cf115dd7bed776512ae7c8c3b6d7a9ad5fc6ee09d27706cef917fda6

See more details on using hashes here.

File details

Details for the file mnemo_db-0.4.0rc2-cp312-cp312-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for mnemo_db-0.4.0rc2-cp312-cp312-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 c5ebc4e416367bce48034d35116f074982099514703e42fe8184723eb3ab3d9e
MD5 5dd46574f03d106f88d0628a25f095f8
BLAKE2b-256 755dbdde4dfa0d4154da070daf97f747a3d37ff4afa0d38c877ac45ed882bf2b

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