Email, chat, and voice input
Receive input through Email Workers, WebSockets and Workers, or Cloudflare Calls, allowing agents to participate in different communication channels.
A platform and SDK for developers building AI agents with persistent state and tools.
Cloudflare Agents is Cloudflare’s platform for building AI agents that receive input, reason with language models, retain state, and act through tools. It brings together the Agents SDK and Cloudflare services for inference, execution, storage, and communication.
Developers can run models with Workers AI or connect to external providers through AI Gateway. Durable Objects and Workflows support stateful, multistep execution, including returning to a model to revise a plan. The platform is suited to applications that coordinate ongoing tasks across APIs, user interactions, and scheduled actions.
Receive input through Email Workers, WebSockets and Workers, or Cloudflare Calls, allowing agents to participate in different communication channels.
Use Workers AI for inference on Cloudflare or AI Gateway to connect to external model providers for planning, reasoning, and content generation.
Combine Durable Objects for state with Workflows for compute to carry out multiple steps and revisit decisions as new variables emerge.
Connect agents to APIs through MCP servers and use Browser Rendering, Vectorize, and D1 to interact with web content, vector data, and databases.
The SDK example demonstrates scheduled methods, callable agent methods, and state updates sent to connected users, supporting recurring tasks and collaborative interactions.
Durable Objects can shut down during inactivity while maintaining WebSocket connections, reducing the cost of agents with long-lived connections.
An office lunch agent can research nominated restaurants, store menus in Vectorize, collect coworkers’ votes, and generate an announcement explaining the selected option.
Developers can use the Agents SDK to build remote MCP servers. The homepage cites Knock’s use of the SDK to ship a server for its customers.
Build agents whose work includes LLM calls, API requests, or human input. Persistent state and execution services support tasks that span these waiting periods.
It describes agents as systems that plan, iterate, and interact with their environment to achieve objectives, extending beyond generating responses to individual prompts.
The homepage describes Workers AI as charging for usage and Workers execution billing in terms of CPU time rather than time spent waiting on I/O. It also highlights Durable Objects WebSocket hibernation as a way to reduce idle connection costs.
The TypeScript example extends Agent, marks methods with @callable(), schedules recurring actions, starts a restaurant research workflow, updates voting state, and calls a Workers AI model to compose the lunch result.
Os dados de tráfego são apenas para referência.
| jul. | 14318 |
|---|---|
| ago. | 11461 |
| set. | 18786 |
As análises de tráfego ainda não estão disponíveis.
openai.github.io
A TypeScript SDK for developers building text, sandbox, and realtime voice agents.
adk.dev
Agent Development Kit (ADK) is an open-source framework for building, evaluating, and deploying conversational and non-conversational AI agents. It supports multi-agent systems, tools, structured workflows, and production runtimes across Python, TypeScript, Go, Java, and Kotlin.
learn.microsoft.com
Microsoft Agent Framework is a developer framework for building, orchestrating, and deploying AI agents and multi-agent workflows. It supports implementations in Python and .NET.
vapi.ai
A developer platform for building, testing, and operating conversational voice AI agents.
lyzr.ai
An enterprise platform for building, deploying, and governing AI agents across workflows.
zaia.app
An agentic operations platform for building and managing AI agents and teams across business channels.