All connectors
VLM Run
Read documents, images, audio and video with VLM Run: extract text and data from PDFs and photos, transcribe speech and describe video. Needs a VLM Run account; usage beyond the free allowance is metered.
Build prompt
Build the connectors at https://clono.sh/connectors/vlm-run and https://clono.sh/connectors/clono.Build a connector from https://clono.sh/connectors/vlm-run/mcp. Use only this address for every call; don't look VLM Run up anywhere else. It signs in with OAuth, not an API key: authorize URL https://gateway.vlm.run/mcp/authorize, token URL https://gateway.vlm.run/mcp/token, dynamic client registration at https://gateway.vlm.run/mcp/register, scopes `openid email profile offline_access`, PKCE with S256, as a public client with no client secret. Then also build the Clono connector from https://clono.sh/connectors/clono/mcp. It needs no sign-in, and it finds more connectors when a task needs one. Tell me when both are built.If your agent says it couldn’t connect, try again. Sign-in sometimes takes a second try.
- 01Open your agent and start a new chat.
- 02Paste the prompt and send it. Your agent builds the connector.
- 03Sign in to VLM Run when your agent asks.