The client wonât add the server
There's no option to add a custom connector
There's no option to add a custom connector
The form is asking for a client ID and secret
The form is asking for a client ID and secret
The URL is rejected or the server never responds
The URL is rejected or the server never responds
https://mcp.usesuperflow.ai/mcp, including the
/mcp path. Some clients silently drop a trailing slash or a missing path
and then fail with a generic connection error.Sign-in fails or keeps looping
The browser window opens and immediately closes
The browser window opens and immediately closes
Signed in, but the client still shows disconnected
Signed in, but the client still shows disconnected
Wrong workspace
Wrong workspace
The consent screen does not list a permission I need
The consent screen does not list a permission I need
npx -y @usesuperflow/mcp login --scope "...". login --help prints every
permission name.The local package wonât start
Not signed in
Not signed in
npx -y @usesuperflow/mcp login, or set SUPERFLOW_API_KEY.The Superflow API key was rejected
The Superflow API key was rejected
SUPERFLOW_API_KEY from
your config and run login instead. A key in the config always wins over a
sign-in.npx is not found, or the package will not start
npx is not found, or the package will not start
npx.Gemini CLI does not show Superflow
Gemini CLI does not show Superflow
gemini mcp list. For a headless gemini -p run,
pass --skip-trust or set GEMINI_CLI_TRUST_WORKSPACE=true.Tools are missing or failing
The assistant says it can't see any Superflow tools
The assistant says it can't see any Superflow tools
/mcp in Claude Code
and the Gemini CLI, Settings â MCP in Cursor) to confirm Superflow is
listed and connected.A tool says forbidden
A tool says forbidden
superflow_get_me), then create a key or connect again with that
permission. The tool reference lists what each tool needs.Tools also canât exceed your own access. If your Superflow account canât do
something, neither can the assistant.A run did not start
A run did not start
An agent run never finishes
An agent run never finishes
superflow_get_run until it completes, so
give it time before asking again. A run cannot be cancelled once started.If it fails outright, the usual causes are the same as for any agent run:
the page isnât publicly reachable, the project has no agents enabled, or the
workspace is out of AI credits.The install check says a project isn't installed
The install check says a project isn't installed
Analytics look empty
Analytics look empty
Disconnecting
Removing the server from your clientâs config stops it calling Superflow, but doesnât revoke the access token it already holds. To revoke the grant itself, disconnect Superflow from the clientâs connected apps screen, or revoke it in Superflow under Settings > Integrations > API keys, in Connected AI apps. For a package sign-in, runnpx -y @usesuperflow/mcp logout. Access dies immediately; anything the assistant already created stays in your workspace.