Cadenya
Connect existing tools to agents and track how their objectives run.
Build application features on a managed agent execution API

Click provides an API for running agents inside an application. Developers configure a model, instructions, and tools, create a stateful session, then submit input and receive messages, events, tool requests, and saved outputs. The service handles agent execution and conversation state while the surrounding application controls the user experience and when work begins.
Agents can use application functions, sandbox tools, and remote MCP connections with configured permissions. Developers supply their own model-provider credentials and remain responsible for end-user authorization and approval rules. Recorded history and saved outputs can persist beyond a run, but interrupted execution is not automatically resumable. Click is currently offered in private preview.
Tried it? Share a useful thought or ask a question.
Good conversations start with a hello.
Sign in to comment