A flexible and extensible AI agent framework for building, testing, and running AI agents with dynamic tool execution.
Phoenix combines AI reasoning with real-world actions through a flexible and extensible architecture.
Agents can call multiple real-world tools like web search, email, shell commands, and more based on context.
Craft, test, and iterate on AI prompts with live feedback in an intuitive interface.
Activate only the tools you need per agent session for optimal performance and security.
Bank-level encryption, compliance controls, and data sovereignty options for sensitive applications.
Easily add or remove tools via clean interfaces and extend functionality with custom components.
Built with Next.js, Tailwind, TypeScript, and the Vercel AI SDK for optimal performance.
Sophisticated memory system allows agents to recall past interactions and maintain context over time.
Create multi-step workflows by chaining tools together for complex task automation.
Phoenix comes with a wide range of pre-built tools that your agents can use to interact with the world.
Build, test, and deploy AI agents in three simple steps.
Craft prompts in the playground and select the tools your agent will have access to.
Iterate on your agent's behavior with live feedback and real-world tool execution.
Integrate your agent into applications or run it as a standalone service with our API.
Phoenix agents can be customized for a wide range of applications across industries.
Get in touch to discuss your use case or request a demo of Phoenix in action.
Contact us for a personalized demo and to discuss your specific requirements.