About Agenta
Agenta is an open source workspace for building, testing and improving AI agents. It can be used by engineers and non-engineers on the same team, and it can run self-hosted or in the cloud.
What it does
Agenta lets you build agents through a chat interface and improve them with feedback from users and reviewers. Templates cover common jobs such as code review, customer support, sales and company knowledge. Agents can be connected to tools such as GitHub and Linear, and they can be reached from Slack or Telegram to run tasks. Sessions and automations let teams see what an agent did and when it runs.
Key features
- Build agents through a chat interface
- Improve agents from user and reviewer feedback
- Templates for code review, support and sales
- Run agent tasks from Slack or Telegram
- Self-host the workspace or use the cloud version
Who it's for
The workspace suits product, engineering and operations teams that want a shared place to build agents. A developer who only needs a prompt test bench may find a smaller tool enough.
Pricing and good to know
The project is open source and can be self-hosted at no license cost. The cloud version and paid plans are described on the pricing page. Agents that call models will use your provider's tokens, so set limits on how often they run.
Highlights
- Build agents through a chat interface
- Improve agents from user and reviewer feedback
- Templates for code review, support and sales
- Run agent tasks from Slack or Telegram
- Self-host the workspace or use the cloud version
