Skip to content
Connectors/Marketplace/
/
GC AI
Last updated

GC AI integration

GC AI is an AI platform for in-house legal teams that covers contract review, legal research, document drafting, and knowledge management. Integrating GC AI with Frontegg lets your application drive GC AI chats and playbook runs, manage files, folders, and projects, maintain playbooks and their checks, work with skills and company profiles, and read usage data on behalf of your users through a single GC AI API key.

Prerequisites

  • A GC AI account with access to app.gc.ai
  • Organization-admin permissions to create an organization API key
  • To create a personal API key, an organization admin must first enable the Allow team to create personal API keys policy

Connect GC AI

Step 1: Open GC AI settings

Go to https://app.gc.ai and sign in. Open the account menu in the bottom-left corner and click Settings.

GC AI account menu with Settings highlighted

Step 2: Open the API settings page

In the settings navigation, select API. This page holds the organization and personal API keys, links to the GC AI API reference, and the usage view.

GC AI settings with the API section highlighted

Step 3: Find the organization API keys section

Scroll to Organization API Keys. Keys created here belong to the organization and are not tied to an individual member.

Organization API Keys section with the Create API Key button highlighted

Step 4: Name the key

Enter a descriptive name such as Frontegg Integration in the Key name (optional) field, then click Create API Key.

Key name field filled in with Frontegg Integration

Step 5: Copy the API key

GC AI shows the full key value exactly once in the API Key Created dialog. Copy it with the copy button and store it in a safe location, then click I've saved my API key.

Keep your credentials secure

The key value cannot be retrieved again after you close the dialog. Never share it or commit it to version control. Anyone holding the key can act on behalf of your GC AI organization.

API Key Created dialog with the key value blurred

Step 6: Verify the key appears in the list

The new key appears under Organization API Keys with its name, a masked preview of the value, and its creation date. You can revoke it from the same row at any time.

Organization API Keys list with the new key

Step 7: Create a personal API key when a tool needs a user identity

GC AI issues two key tiers, and the connector accepts either one. An organization key (prefix gcai_) carries no individual user, while a personal key (prefix u:gcai_) acts as the member who created it. Some operations require one tier and reject the other, and each tool states its own requirement in its description.

To create a personal key, turn on Allow team to create personal API keys under Personal API Keys as an organization admin, then have the member enter a key name and click Create API Key in that section. The key value is shown once, exactly as in step 5.

Which tier to use

Personal keys are required for user-scoped operations such as starring a project, duplicating a playbook, editing playbook checks, and reading or updating your own profile. Organization keys are required for organization-scoped operations such as updating a company profile and reading credit usage or the usage summary. Neither tier covers the whole API surface.

Personal API Keys section with the policy toggle and Create API Key button highlighted

Configure the Frontegg portal

Once you have your GC AI API key from the steps above, enter it in the integration configuration page of the Frontegg portal:

  1. Open the Frontegg portal and navigate to [ENVIRONMENT] → Integrations → GC AI.
  2. Paste the key into the API Key field.
  3. Click Save.

Keep your credentials secure

Never share or commit your GC AI API key to version control. Revoke a key from Settings → API as soon as it is no longer needed.

Provider limitations

  • Authentication — The GC AI External API authenticates with an API key. GC AI runs an OAuth 2.0 authorization server, but it exposes only identity scopes and no product scopes, so OAuth is not used by this connector and there are no scopes to select.
  • Key tiers — The tier gate applies in both directions: some tools refuse an organization key and others refuse a personal one. GC AI checks the tier before it looks up the resource, so a mismatched key returns a tier error rather than a not-found error.
  • File uploads — The connector uploads a file in a single request. GC AI's resumable multi-part upload flow is not exposed.
  • MCP credentials — Minting OAuth client secrets for the GC AI MCP server is not exposed through the connector. Manage those under Settings → API → MCP credentials.
  • Rate limits — Inference endpoints such as chat completions and playbook runs are limited to 60 requests per minute per organization and 20 per API key. Other endpoints are limited to 600 requests per minute per organization and 120 per API key. Exceeding a limit returns 429 Too Many Requests with a Retry-After header.

Additional resources