midas-memory-mcp
The MCP server for Midas — local-first, source-traceable agent memory with no LLM at ingest. Local embeddings + ranking only: ingest is $0, nothing leaves your machine, and every recalled memory points back to its source turn (no LLM-rewritten facts).
This package is a thin launcher. It installs the Midas SDK (midas-memory) with the MCP and
local-embedding extras and puts a midas-memory-mcp command on your PATH that any MCP client
(Claude Code, Cursor, Codex, Windsurf, …) can run.
Install & run
uvx midas-memory-mcp # run once, nothing to install
# or put the command on your PATH:
pipx install midas-memory-mcp
pip install midas-memory-mcp
Wire it into a client (example: Claude Code)
claude mcp add midas -s user -- midas-memory-mcp
On connect, Midas injects a short memory policy into the agent and starts remembering on its own — it scores importance locally (no LLM), keeps what matters, and drops trivia and duplicates.
Optional environment variables: MIDAS_MCP_DB (SQLite path to persist across restarts),
MIDAS_MCP_MAX_RECORDS, MIDAS_MCP_MIN_IMPORTANCE, MIDAS_MCP_EMBEDDER.
For Cursor / Codex / Windsurf / Claude Desktop config, the Python SDK, and reproducible benchmarks, see the main repo: https://github.com/vornicx/Midas
Tip: if you installed the SDK directly (
uv tool install "midas-memory[mcp,local]"), the same server is also available as the shortermidas-mcpcommand.
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for midas-memory-mcp 0.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| midas_memory_mcp-0.0.1.tar.gz | 2.3 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| midas_memory_mcp-0.0.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 5.0 kB
Release files / midas_memory_mcp-0.0.1.tar.gz
| Download URL | midas_memory_mcp-0.0.1.tar.gz |
|---|---|
| Size | 2.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
926c6f9c7b0341a2b43516b377b8a9efaf868cd4d3de08cd455ef8c8d2859b32
|
|
BLAKE2b-256 checksum How to use checksums |
c1c55f08b140046ca3832c8f288dbfaa8026d545612c3d31466bcb5b441887b9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.18 {"installer":{"name":"uv","version":"0.11.18","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Pop!_OS","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / midas_memory_mcp-0.0.1-py3-none-any.whl
| Download URL | midas_memory_mcp-0.0.1-py3-none-any.whl |
|---|---|
| Size | 2.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
8df0188d9c2e2bcab30e215e9cd90f788f35f115f8f1638b0ae1d0419001bb7b
|
|
BLAKE2b-256 checksum How to use checksums |
37a2dd127daf343e5d854ceed0a87dee1ea543711a45c840473b83dce875f6a4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.18 {"installer":{"name":"uv","version":"0.11.18","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Pop!_OS","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for midas-memory-mcp 0.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
