ForgingStack Mandate Check

io.github.jdkabattlesPublished by GitHub account jdkabattlesv0.2.1Updated Nov 4, 2026

Before your agent acts, check its mandate. One call, one receipt.

The publisher's own server: check.tattoo: tattoo size and placement guide

VerifiedSTDIODesktop onlySecurity & MonitoringAI & ML

Overview

AI-generated overview

Checks an AI agent's proposed action against its mandate and returns a pass/fail decision with a hash-chained receipt.

What you can do
  • Check a proposed agent action against a mandate and get a pass/fail decision with a reason and receipt
  • Retrieve a past mandate-check receipt by its id
  • Read the remaining credit balance for the configured API key
  • Verification covers agent match, principal match, mandate expiry and whether the action is in the allowed list
When to use it
Use it when an agent should confirm that an action falls within a defined mandate before acting, and when you want a verifiable receipt of that check for later audit.
Requirements
Runs locally over stdio, launched with uvx from the PyPI package. Needs a FORGINGSTACK_API_KEY, obtained by signing up to the hosted service, and network access to the API base URL (configurable via FORGINGSTACK_API_BASE). Credits are required; signup includes 100 starter credits.
Before you install
Each check costs 1 credit, including failed decisions, so calls spend money; only idempotent replays are free. The API key is shown once at signup and stored only as a hash, so it must be saved immediately. The decision logic, receipt chain and billing run on a third-party hosted service, so mandate data and action details are sent to it. It checks mandate fit only, not whether an action is wise, legal or safe.

Installation

In SourceWeft

  1. Open ForgingStack Mandate Check in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

ForgingStack Mandate Check

Before your agent acts, check its mandate. One call, one receipt.

mcp-name: io.github.jdkabattles/forgingstack-mandate-check

This is the public client for Mandate Check. The service itself — the decision logic, the receipt chain, the key store, and billing — runs on ForgingStack's private service and is not in this repo. Copying this repo does not give you the service: you still need an API key and credits, from signup below.

Live API: https://forgingstack-mandate-check.jkbattles.workers.dev

What it checks, in order

  1. Mandate agent matches the calling agent
  2. Mandate principal matches the principal the agent acts for
  3. Mandate expiry is parseable and in the future
  4. Action is in the mandate's allowed actions

Pass or fail, you get a reason and a hash-chained receipt.

What it does not do: decide if an action is wise, legal, or safe. Mandate fit only.

Signup (v1)

bash
curl -X POST https://forgingstack-mandate-check.jkbattles.workers.dev/v1/signup \  -H "Content-Type: application/json" \  -d '{"email":"[email protected]"}'

Returns an API key once and 100 starter credits. Save the key immediately — it is stored only as a hash and cannot be shown again. Email verification is pending, and credit-pack billing is not live yet, so do not expect a purchase flow in v1.

Agent sub-keys

A signed-up human creates a key for each agent, capped and revocable:

bash
curl -X POST https://forgingstack-mandate-check.jkbattles.workers.dev/v1/keys \  -H "X-API-Key: YOUR_PRIMARY_KEY" -H "Content-Type: application/json" \  -d '{"agent_name":"my-agent","credits":20}'

List with GET /v1/keys, revoke with DELETE /v1/keys/{id} — revoking returns unused credits to the primary account.

Use it (MCP, stdio)

Set the key, then run the MCP server in this repo:

bash
export FORGINGSTACK_API_KEY="your-key-here"python -m app.mcp_server

Tools: check_mandate, get_credits, get_receipt. See SKILL.md for when to use it and when not to.

Price

1 credit per check call. A failed decision still costs 1 credit — the check was performed. An idempotent replay (same Idempotency-Key) is free and returns the original receipt.

Files in this repo

  • app/mcp_server.py — MCP wrapper client (stdio, stdlib only)
  • SKILL.md — when to use it, how to read pass/fail
  • agent-card.json — A2A-style discovery card
  • server.json — MCP Registry metadata (draft until PyPI publish)
  • pyproject.toml — package metadata for the MCP client

Source: README.md at commit 315cc45

Tools

3

check_mandate

Check an agent action against its mandate. Returns a pass/fail decision and a receipt. Costs 1 credit per call; an idempotency key replay is free.

Parametersactionrequiredmandaterequiredagent_idrequiredprincipalrequiredidempotency_key

get_credits

Remaining credit balance for the configured API key.

Takes no parameters.

get_receipt

Fetch a past mandate-check receipt by id.

Parametersreceipt_idrequired

Version history

1
  1. v0.2.1LatestNov 4, 2026

Score

80% 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 days25 pts
  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.

Other Check integrations