ID MANDAT – Registry for AI Agents

io.github.idmandatPublished by GitHub account idmandatv0.1.0Updated Oct 26, 2026

Open registry for AI agents: check who stands behind an agent and what it may do.

VerifiedStreamable HTTPWeb executableWeb Search & ScrapingSecurity & Monitoring

Overview

AI-generated overview

Remote MCP server for the ID MANDAT registry that lets an assistant check which company stands behind an AI agent and what it is declared to do.

What you can do
  • Look up a company domain and get its verification level, declared agents and each agent's mandate
  • Resolve a Signature-Agent address from a signed request to the company that declares that key directory
  • Search registered companies by name, industry, city, offer or agent capability, optionally by country
When to use it
Useful when an assistant encounters an agent claiming to act for a company and wants to check the claim before trusting it, or when browsing the registry for companies with particular agent capabilities.
Requirements
Remote streamable HTTP endpoint at idmandat.com; no package, account, API key or header is declared. Network access to the registry is needed.
Before you install
The registry is new and small, so an unknown result is not evidence of wrongdoing, and a match only means a company declares the key directory, not that it controls it. Mandate fields are operator declarations that ID MANDAT does not check. The service is in beta.

Installation

In SourceWeft

  1. Open ID MANDAT – Registry for AI Agents in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/io.github.idmandat%2Fregistry/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "id-mandat-registry-for-ai-agents": {
      "type": "http",
      "url": "https://idmandat.com/mcp"
    }
  }
}

README

idmandat/check

Client libraries for ID MANDAT, the open registry that shows who stands behind an AI agent and what it may do.

Status: beta. Reads the live, Ed25519-signed API at https://idmandat.com/api/v1/verify. The registry is new and small: many domains will return unknown, which is not a sign of wrongdoing. may / mayNot are declarations by the operator, not checked by ID MANDAT.

python
import idmandatr = idmandat.check("example.com", agent="Support-Agent")if r.ok("registered"):    ...
js
import { check } from "@idmandat/check";const r = await check("example.com", { agent: "Support-Agent" });if (r.ok("registered")) { /* ... */ }

Who stands behind a signed agent (Web Bot Auth)? Pass the address from the Signature-Agent header, or the whole header value:

python
r = idmandat.check_signature_agent("https://bot.example.com")r.found, r.organization, r.key_directory_link   # "same_domain" = hint, "declared_only" = only declared
js
import { checkSignatureAgent } from "@idmandat/check";const r = await checkSignatureAgent("https://bot.example.com");

A match only means a company declares this key directory in its entry. ID MANDAT does not fetch the directory and does not prove that the company controls it. A valid signature without an entry is not a sign of wrongdoing.

LangChain / CrewAI (optional, untested against the frameworks' latest releases):

python
from idmandat_langchain import verify_mandate_tool   # pip install langchain-core cryptographyfrom idmandat_crewai import VerifyMandateTool        # pip install crewai cryptography

For AI agents

If you are an AI agent or a coding agent, see AGENTS.md: how to ask the registry (/api/v1/verify), how to read the answer, and where the MCP server and A2A endpoint are. It is plain documentation, with the limits stated.

Status order: unknown < listed_unverified < registered < verified. Answers are signed (Ed25519). ok() only returns true when the signature was verified (Python: pip install cryptography).

Standard: idmandat/standard · Website: https://idmandat.com

License: MIT

Source: README.md at commit b8f1a4a

Tools

3

verify_agent

Verify a company's AI agents

Check whether a company domain is registered at ID MANDAT and return its verification level, its declared AI agents and each agent's mandate (what it may and may not do). Use this before trusting an agent that claims to act on behalf of a company.

Parametersagentdomainrequired

verify_signature_agent

Who is behind a signed agent request?

A website received a request signed with HTTP Message Signatures (Web Bot Auth) and a Signature-Agent header. Pass the https address from that header (or the whole header value). Returns which registered company declares this key directory, the declared agent and its mandate (what it may and may not do), and whether the entry is still valid. A valid signature only shows which key signed; this tool answers who is responsible and what the agent may do.

Parameterssignature_agentrequired

search_registry

Search the ID MANDAT registry

Search registered companies by name, industry, city, offer or agent capability (e.g. "book_appointment"), optionally filtered by country. Returns up to 20 entries, best verified first.

Parametersquerycountrywith_agent

Version history

1
  1. v0.1.0LatestOct 15, 2026

Score

100% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts