Troubleshooting

Fix server URL, sign-in loop, organization, authorization, permission, and environment issues.

Most MCP connection issues come from one of three causes: the wrong server URL, the wrong signed-in account, or the wrong organization. Start with those before changing client config.

Wrong server URL

Confirm the client is using https://app.autonomyqa.com/api/mcp or the matching host for your dedicated deployment, and that the URL ends with /api/mcp.

Clear the client connection, sign out in the browser, and authorize again. If the loop continues, check that the browser can reach the same Autonomy app host used by the MCP URL.

Missing organization

Use an account that belongs to the organization. If access was just granted, refresh the browser authorization flow.

Denied authorization

The organization may block the client or your role may not include MCP access. Ask an organization admin to review access.

Environment mismatch

Verify that the plan and environment belong to the same organization and deployment tier.

Client cache

Some clients cache authorization state. If the browser flow succeeds but tools still fail, remove the Autonomy MCP entry from the client, restart the client, and add the server again.