§ A — Free integrations

Agent surfaces.

ROBOT.md works with any MCP-aware agent. Five surfaces are tested and shipped.

§ B — Managed offering

Compliance in one day. Not one month.

Give Compliance-bot your robot repo. Get back a signed FRIA, IFU, safety-benchmark, and EU-register filing — all five RCAN §22–26 packets — without reading a single EU AI Act article yourself.

§ B.1 — Compliance-bot

Automated compliance filing for regulated robots.

The RCAN §22–26 compliance stack is already live. Compliance-bot is the hosted agent that automates it.

Input
Your robot repo
Point Compliance-bot at a GitHub repo containing a valid ROBOT.md with an issued RRN. Nothing else required.
Analysis
Manifest audit
The agent reads the manifest, cross-checks capabilities against RCAN §22 FRIA criteria, and drafts the attestation text. Flags gaps for human review before filing.
Filing
5 packets signed + sent
FRIA · safety-benchmark · IFU Art. 13(3) · incident-report Art. 72 · EU-register Art. 49 — all signed with your robot's key and routed to RRF's live endpoints.
Output
Signed receipts
You get a compliance bundle: signed envelopes, RRF confirmation hashes, and a human-readable summary ready for your legal team or regulator.
§ B.2 — Open-core model

OSS agent SDK. Hosted convenience layer.

The compliance tooling is open-source. The hosted agent is the convenience layer — no infrastructure to run, no SDK to integrate.

Open source · Apache-2.0
robot-md-gateway

The reference enforcement gateway: receives signed action requests, verifies, and dispatches to drivers. Self-host on your robot or your own cloud.

  • Full robot-md-mcp surface — 4 resources + 2 tools (frontmatter, capabilities, safety, body · validate, render)
  • Default-deny tier gate; hardened systemd unit
  • Audit log; RCAN-signed envelopes for all dispatch events
  • All §22–26 filing commands available as CLI primitives
  • Provision with /enable-dispatch from Claude Code
github.com/RobotRegistryFoundation/robot-md-gateway →
Managed Agent · hosted
Compliance-bot

The hosted Managed Agent. Give it a repo, get back a full compliance bundle — no infrastructure, no CLI, no reading of RCAN section numbers.

  • Accepts a GitHub repo URL or a ROBOT.md file upload
  • Runs a pre-flight audit and flags gaps before filing
  • Files all 5 §22–26 packets in one session
  • Returns signed receipts + human-readable compliance bundle
  • Powered by Anthropic Managed Agents (see §B.3 below)
Join the waitlist ↓
§ B.3 — Anthropic surface alignment

Built on Anthropic Managed Agents.

Compliance-bot is the first robot-md product built on Anthropic's just-announced Managed Agents platform — composable cloud-hosted agents with a public API.

Anthropic · Managed Agents (beta)

Composable APIs for cloud-hosted agents. robot-md's commercial wedge runs here.

Anthropic's Managed Agents platform provides the infrastructure for cloud-hosted, composable agent workflows. Compliance-bot is the robot-md flagship built on this surface: it receives a robot repo, orchestrates a multi-step compliance filing workflow, and returns signed attestation packets — all without the operator running any local infrastructure.

The open-source robot-md-gateway is the reference enforcement gateway — self-hosted for developers who want full control. Compliance-bot is the managed convenience layer for operators who want the outcome without the infrastructure.

Anthropic: Product development in the agentic era →
Waitlist · Compliance-bot

Get early access.

Compliance-bot is in private preview. Leave your email and we'll contact you when it's ready — or when we need a test case. No spam. No sales cadence. One email when the beta opens.

Your email is only used to contact you about Compliance-bot. No third-party sharing. Unsubscribe in one reply.

01 — Wire it

Do the cookbook.

Install robot-md, author a ROBOT.md, register with RRF, run a skill, and walk away with a signed audit bundle. About ten minutes, end to end.

Open the cookbook →
02 — Read it

Understand the format.

The ROBOT.md spec defines every required and optional field, the RCAN protocol conformance rules, and the compliance packet filing requirements. Normative text, schema, and examples.

Read the spec →
03 — Want this managed?

Compliance-bot waitlist.

Give your robot repo to Compliance-bot and get back a signed FRIA + IFU + safety-benchmark + EU-register filing in one day. Powered by Anthropic Managed Agents.

Join the waitlist →
Where safety is actually enforced.
Physical safety is enforced at Layer 3 (robot-md-gateway) or Layer 4 (a runtime that embeds it, e.g., OpenCastor). Declaration alone (Layer 1) does not enforce safety. Agent host alone (Layer 2) is not the safety boundary. If a deployment lacks Layer 3, no safety claim attaches to it.