Click

Managed agents API for running serverless agents with any model

Verified clicks, 30 days: 0
Votes: 0
Score: 15.0 pts
Comments: 0

About

Click is a managed agents API that lets you run serverless agents with any language model and your own tools. You send work via an API call, and the agent handles it by browsing, searching, running code, or using tools you've connected. Results stream back to your app in real time, or you can receive a webhook when the work completes.

You bring your own model keys and choose which tools the agent can access for each session. Connect MCP servers, HTTP endpoints, or custom application tools. The platform provides built-in capabilities like a sandbox for running code, a browser for web interaction, and web search, which you can use or replace with your own.

Click runs on a pay-as-you-go model based on concurrent sessions and usage of managed services like sandbox hours and browser hours. The free plan includes 2 concurrent sessions, 100 MB storage, and limited managed services. The Pro plan scales to 50 concurrent sessions with higher quotas for sandbox and browser hours.

Click lets you use any model with your own keys and bring your own tools via MCP servers or HTTP endpoints. Every plan includes the full platform with streaming results and webhooks, so you only pay for capacity and managed services, not for features.

Problem: Building agents into applications requires managing model integrations, tool connections, session state, and async workflows, which adds complexity to your codebase.
Solution: Click handles agent orchestration, tool execution, and result streaming through a simple API, so you focus on your application logic rather than agent infrastructure.

Updated today