Bring your own key
Release Autopilot's agents use an AI model to read pull requests. Bring your own key is optional: a project that does not set it up uses LaunchDarkly's model account (the product calls this "LaunchDarkly managed inference"). No MCP tool reads or writes inference settings, so this is a browser step.
For each project, give the user /projects/{projectKey}/settings/automations on their LaunchDarkly instance (for example https://app.launchdarkly.com, or their EU or federal URL). This is the project's Settings > Automations page. If the link opens the general project settings page instead, the account uses the older settings pages, and they use /projects/{projectKey}/settings/vega. In the Inference section they:
- Select bring your own key. If the account has no Anthropic or OpenAI key yet, the page links to the integration form to add one. Keys are account-wide, so later projects can reuse them.
- Keep or change the suggested default and small models.
- Save. LaunchDarkly checks the key with the provider before it stores the settings, and shows the cause if the check fails.
Record the result as reported by the user. It cannot be verified.