# Sovereign Agent Logbook

> Sovereign Agent Logbook is an open-source MCP server for security. Hash-chained, tamper-evident action log for AI agents, stored on your own server. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, self-hosted, Linux
- **Repository:** https://github.com/delfen-com/sovereign-agent-logbook
- **MCP Registry name:** io.github.Delfen-com/sovereign-agent-logbook
- **Categories:** [security](https://ai-powered.com/categories/security), [legal and government](https://ai-powered.com/categories/legal-and-government)
- **Listing updated:** 2026-11-18

## About

Each logged event carries the hash of the previous one, so changing, inserting, removing or reordering a line makes logbook verify report exactly where the chain breaks. An entry records who acted, what, why, who approved it and the outcome, and every log_event call must name the agent, the model and the action so an auditor can see which model version did what. Values that look like passwords, tokens, API keys, IBANs or BSNs are masked before hashing.

The author frames it around EU AI Act record-keeping and human-oversight duties, noting that the Digital Omnibus on AI moved most high-risk deadlines to 2 December 2027, and says it helps with a verified timeline but not every article. A public test blocks outbound sockets and still completes a write, verify and export cycle. Exports come as JSON or CSV.

The log format is a separate Apache-2.0 spec. The README calls v0.1 a working skeleton. Run it over stdio from pip or as a Docker service; a hosted EU version at 29 euros a month is planned.

## Features

- **Hash-chained log:** Tampering is detected and located by verification.
- **Secret masking:** Passwords, tokens, API keys, IBANs and BSNs are masked before hashing.
- **No outbound network:** A public test proves a full cycle with sockets blocked.
- **Audit export:** JSON and CSV readable without the tool.
- **Open format spec:** The log format is documented separately under Apache-2.0.
- **Skills pack:** Teaches agents when to log and when to ask a human.

## Good for

- Keep an auditable record of an invoice-paying agent's actions
- Reconstruct which tool call did what after an incident
- Hand an auditor a CSV export of agent activity
- Log scripted jobs through the logbook command line

## Install

### Claude Code (Terminal)

```sh
claude mcp add sovereign-agent-logbook -- uvx sovereign-agent-logbook
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "sovereign-agent-logbook": {
      "command": "uvx",
      "args": [
        "sovereign-agent-logbook"
      ]
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "sovereign-agent-logbook": {
      "command": "uvx",
      "args": [
        "sovereign-agent-logbook"
      ]
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "sovereign-agent-logbook": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "sovereign-agent-logbook"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "sovereign-agent-logbook": {
      "command": "uvx",
      "args": [
        "sovereign-agent-logbook"
      ]
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "sovereign-agent-logbook": {
      "command": "uvx",
      "args": [
        "sovereign-agent-logbook"
      ]
    }
  }
}
```

## Tools

- `log_event`: Record an agent action with who, what, why, approval and outcome.
- `verify_logbook`: Verify the hash chain and report where it breaks.
- `recent_events`: Read the most recent events.
- `export_logbook`: Export the logbook as JSON or CSV.

## FAQ

### Does it send data anywhere?

No; it writes to a file on your server and a test checks it makes no outbound connections.

### Which models work?

Any, because it speaks MCP.

### What Python version is needed?

3.11 or newer.

### Is there a hosted version?

One is planned at 29 euros a month with a 30-day trial.

### Is Sovereign Agent Logbook free?

Yes. Sovereign Agent Logbook is open source, so you can use and self-host it for free.

### How do I add Sovereign Agent Logbook to Claude Code?

Run this command in your terminal: claude mcp add sovereign-agent-logbook -- uvx sovereign-agent-logbook

### What are the best alternatives to Sovereign Agent Logbook?

The closest alternatives to Sovereign Agent Logbook are REA, AiSOC, redamon, x64dbg-mcp-server and Agent Roast.

## Where this comes from

- Description: written from the project's README.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [AiSOC](https://ai-powered.com/mcp/aisoc): AiSOC is an open-source MCP server for ai models and llm tooling.
- [redamon](https://ai-powered.com/mcp/redamon): redamon is an open-source MCP server for ai models and llm tooling.
- [x64dbg-mcp-server](https://ai-powered.com/mcp/x64dbg-mcp-server): x64dbg-mcp-server is an open-source MCP server for ai models and llm tooling.
- [Agent Roast](https://ai-powered.com/mcp/agent-roast): Agent Roast is an open-source MCP server for developer tools.
- [Midplane](https://ai-powered.com/mcp/midplane): Midplane is an open-source MCP server for databases.
- [Leash](https://ai-powered.com/mcp/leash): Leash is a free MCP server for security.
- [Data Ny](https://ai-powered.com/mcp/data-ny): Data Ny is a freemium MCP server for legal and government.

---
Source: https://ai-powered.com/mcp/sovereign-agent-logbook
