inspeximus: agent memory with provenance, GDPR erasure and a tamper-evident audit trail MCP Server

What the code does

We read the source, 22 h ago · rules 3dff92dd89df

A value the model can set ends up inside a file or shell call. That is not a flaw by itself — for a terminal server it is the job — but it is where things go wrong when it is not.

            proc = subprocess.run([sys.executable, "-c", code], cwd=tmp, capture_output=True,

What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.

    m = Inspeximus(path=os.path.join(d, name))

    return subprocess.run(

                for k, n in eval(kinds):  # noqa: S307 - our own repr of a sorted list of pairs

                <td><span class="mk no">delete + legal-hold, not verifiable</span></td>

    env = dict(os.environ)

Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.

Connect this server

This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.

claude mcp add inspeximus -- uvx inspeximus

{
  "mcpServers": {
    "inspeximus": {
      "args": [
        "inspeximus"
      ],
      "command": "uvx"
    }
  }
}

[mcp_servers.inspeximus]
command = "uvx"
args = ["inspeximus"]

{
  "mcpServers": {
    "inspeximus": {
      "args": [
        "inspeximus"
      ],
      "command": "uvx"
    }
  }
}

{
  "mcpServers": {
    "inspeximus": {
      "args": [
        "inspeximus"
      ],
      "command": "uvx"
    }
  }
}

This one needs environment variables set before it will start: INSPEXIMUS_PATH (Path to the JSON memory file the server persists to (defaults to ./inspeximus_memory.json).), INSPEXIMUS_ECHO_GUARD (1 (default) blocks a restated retired value from resurrecting a corrected fact; 0 = legacy keyed supersession.), INSPEXIMUS_EMBED_URL (Optional OpenAI-compatible /embeddings endpoint for SEMANTIC recall (e.g. http://localhost:11434/v1/embeddings for local Ollama). Unset = deterministic lexical recall.), INSPEXIMUS_EMBED_MODEL (Embedding model id used with INSPEXIMUS_EMBED_URL (default: text-embedding-3-small; for Ollama use e.g. nomic-embed-text).). The author declared them in the registry entry; get the values from the project itself.