Skip to content
English
  • There are no suggestions because the search field is empty.

Developer tools - integrate with CyberHQ.ai

Summary

CyberHQ.ai exposes three ways to integrate programmatically: a versioned REST API, an MCP server for AI-assisted tools like Claude Desktop, Claude Code, and Cursor, and signed outbound webhooks. All three are built on the same foundation, a scoped API key you create yourself, so understanding that key once covers every integration path.


The three integration surfaces

  • REST API: a versioned HTTP API (/api/public/v1/...) covering signals, events, users, assets, compliance, policies, incidents, and remediation. Some areas need a higher plan: signals, compliance frameworks and attestations, incidents, remediation, and changing policies need Professional. Compliance programs come with Ultimate or the Full compliance add-on, and also have to be turned on for your organization. A request to an area your plan doesn't include is refused with a 403 and the code upgrade_required. See Explore the REST API reference.
  • MCP server: six read-only tools an MCP-compatible client can call directly, for searching signals and exposure chains and looking up users and vendors. It reads your data; it can't change anything. The MCP tools are part of Iris, the built-in AI assistant, so they need the Ultimate plan: on Starter and Professional a key still connects, but the client sees an empty tool list rather than an error. Five of the six tools identify people, so they also need your organization's Display mode set to Named. See Connect an MCP client.
  • Outbound webhooks: the platform pushes events to a URL you control, signed so you can verify they genuinely came from CyberHQ.ai. See Verify webhook signatures.

One credential model for all of it

Every REST and MCP request authenticates with an API key you create under Settings → Integrations → API keys, which needs the Manage Tenant permission (see Connect an integration for the creation steps). A key carries:

  • A set of granular scopes: exactly what it's allowed to read or write, nothing more. Scopes are fixed when you create the key; to change them, create a new key. See Choose the right API key scopes.
  • A rate limit: a per-minute request cap, 120 by default. You can set a different limit, up to 6,000, when you create the key. See Authenticate your API requests.
  • An optional expiry date: the key stops working at the end of that day. Without one, it keeps working until you Rotate or Revoke it, so revoke keys you no longer use.

This is a genuinely separate credential from your own dashboard login: a key's access is defined entirely by its scopes, never by whichever role happened to create it.

Where to start

  1. If you haven't created a key yet, start with Connect an integration.
  2. Read Authenticate your API requests to understand how requests are authenticated and what a rejected request looks like.
  3. Pick the integration path that fits what you're building, REST, MCP, or webhooks, from the three articles linked above.

A note on stability

The REST API's contract is stable: it's documented and versioned independently of the platform's internal interfaces, specifically so an integration built against it doesn't break as the product evolves. The OpenAPI document itself is public and unauthenticated, so you can inspect the full contract before writing any code.

Related