mcpField
W
Warrant

Warrant

Local
github.comAI & Agents5Updated today
RepositoryVisit website

About Warrant

Warrant is an MCP server for agents to file signed, hash-addressed decision records and verify a store's integrity. It suits workflows that need recorded reasoning and offline re-execution of supported checks.

Key features

  1. 01File propose, accept, reject, and supersede decisions
  2. 02Sign and hash-address decision records
  3. 03Verify record hashes, signatures, and links
  4. 04Show reasons and the verified chain
  5. 05Re-execute ski@v1 checks

Use cases

  1. 01Keep an agent's decisions and rationale
  2. 02Check a store for tampering
  3. 03Re-run executable decision reasons offline

MCP support

The stdio MCP server gives clients tools to file decisions, verify a store, and inspect reasons. It runs operations through the warrant CLI as a subprocess and does not wrap another MCP server.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "s0fractal-warrant": {
      "command": "uvx",
      "args": [
        "warrant-verify"
      ],
      "env": {
        "WARRANT_STORE": "<WARRANT_STORE>",
        "WARRANT_KEY": "<WARRANT_KEY>",
        "WARRANT_ACTOR": "<WARRANT_ACTOR>"
      }
    }
  }
}
Run commanduvx warrant-verify
Environment variables to set
WARRANT_STOREAbsolute path to the warrant store this server files into and verifies. Defaults to .warrants relative to the server process's working directory, which an MCP host does not guarantee.
WARRANT_KEYSecretEd25519 signing key. If unset, one is generated next to the store on the first filing and the result says so; any process that can read it can sign as this actor.
WARRANT_ACTORActor id written into filed warrants. Defaults to mcp-client@<hostname>.

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypiwarrant-verify0.9.0

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

The maintainer has not added details yet. Check the README in the repository.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.