Skip to main content

Iron Brick Developer Portal

One portal for APIs, events, and AI agents.

Find, test, and govern every integration in one place, built for civilian federal and homeland security programs. Get sandbox keys in minutes, then deploy behind your agency's API gateway.

Two calls to your first case
curl -s -u "$CLIENT_ID:$CLIENT_SECRET" \
  -d grant_type=client_credentials \
  https://dev.ironbrick.us/oauth/token

curl -s -H "Authorization: Bearer $TOKEN" \
  https://dev.ironbrick.us/sandbox/cases/v1/cases?status=in_review

The problem

Integrations are scattered, and agents have no home

Developers look for APIs in one portal and event streams in another, and AI agents aren't cataloged anywhere. Onboarding takes weeks. Internal teams build services that already exist. AI engineers ship agents without governance because no catalog is built for agents.

Scattered

APIs, events, and agents live in different tools, or nowhere at all.

Slow onboarding

Weeks of tickets and email to get credentials and a working example.

Redundant services

Teams rebuild what already exists because they can't find it.

Ungoverned agents

No owner, scope, or oversight on record for what an agent can touch.

One catalog. Three kinds of assets. One set of rules.

Every asset has an owner, documentation, a sandbox, and the same identity, scopes, and audit trail.

Minutes, not weeksSelf-service keys and a working sandbox call
Reuse before you buildSearch everything that already exists
Governed by defaultAgents get an owner, scopes, and oversight before they ship

Robust APIs

Versioned, contract-first APIs for cases, applications, documents, records requests, events, and audit, modeled on how agency systems of record actually work.

Browse the catalog

Comprehensive docs

OpenAPI 3.1 downloads, request and response examples, error formats, rate limits, and step-by-step guides from first token to production.

Read the docs

Integrations & AI agents

Model Context Protocol tools let approved AI assistants search and summarize records with citations, under the caller's own permissions.

See the agent tools

Start building in three steps

The sandbox is free and returns synthetic data. No real or personal information is ever involved.

1

Register

Create a free developer account and verify your work email.

Create an account
2

Create an app

Choose the scopes you need and copy your client ID and secret.

Manage applications
3

Call the sandbox

Exchange credentials for a token, validate it, and make your first call.

Follow the quickstart

Built for civilian federal and homeland security programs

Designed to drop into the controls your agency already runs.

Identity first

OAuth 2.0 with short-lived tokens and least-privilege scopes. Agency deployments federate with your IdP, including PIV/CAC-backed sign-in.

Every call audited

Who, on whose behalf, which record, and the outcome, streamed to your SOC/SIEM in JSON or CEF.

Your boundary

Deploy inside your FedRAMP-authorized cloud or on premises, behind your API gateway. Designed to support NIST SP 800-53 controls.

AI with receipts

Agent tools are read-only by default, cite their sources, and require a human to approve any change.

Iron Brick reference APIs are not an official U.S. government system. Agency names are not used to imply endorsement.

Case study · FDA

From FedEx boxes to near real-time

Medical device manufacturers mailed paper to FDA, and staff on both sides processed it by hand. An API-driven manufacturer portal replaced that with registration, security validation, virus checking, document-to-database conversion, and Tableau monitoring.

Read the case study
  • Paper and courier steps removed
  • Manual entry replaced by automated checks and conversion
  • Near real-time audits and device updates
  • Live dashboards for volume and exceptions
  • Delivered in under a year, under close agency oversight

Planning a modernization or AI pilot?

Talk with an Iron Brick architect about adapting these APIs to your systems of record.

Talk to an Architect