The open-source workspace for building and running agents
Build agents that automate your work by chatting with them. Share them with your team, connect them to the apps you use, and run them in the background.
Documentation • Website • Agenta Cloud
What is Agenta?
Agenta is an open-source workspace where you build specialized agents that automate and augment your work.
You build agents by chatting with them. You describe the work, connect the apps they need, and improve them through feedback.
You can work with your agents directly in chat and share them with your team.
For recurring work, you can build background agents. These agents run on a schedule or when an event occurs.
Why use Agenta?
Use your Claude or ChatGPT subscription
When you self-host Agenta, you can run agents locally with your existing Claude or ChatGPT subscription. You do not need to move every task to metered API billing.
Choose your harness and model
Switch harnesses and models without rebuilding your agent. Agenta already supports almost any model, whether self-hosted or accessed through an API. Agenta supports Claude Code, Pi, and Codex as harnesses today, with more harnesses planned.
Build with open agent standards
Define your agent with AGENTS.md, skills, and MCP servers. You can bring skills and MCP servers from the agent ecosystem into Agenta.
Make your agents more reliable over time
Agenta traces every run and keeps a version history of each agent configuration. Use this history to understand failures, compare changes, and improve your agents over time.
Features
Workspaces for you and your agents. Work with your agent on files in a shared workspace. Together, you can write documents, organize research, or maintain a wiki.
Human approval and permissions. Specify permissions for each tool. Choose which actions background agents can run automatically, which need your approval, and which are blocked.
Background agents. Run agents on a schedule or start them when an event occurs in a connected app.
Tracing, usage, and cost. Inspect every model and tool call. Track model requests, token usage, and estimated costs for each agent.
Team access. The open-source version lets you share agents with your team and control access by role.
Integrations. Connect your agents to the applications you use through MCP, or integrate with more than 1,000 apps through Composio, including Gmail, Slack, Notion, and GitHub.
Get started
Try Agenta Cloud
The fastest way to try Agenta.
Self-host Agenta
Paste this into your agent and it will walk you through setup and testing:
1. Install the Agenta self-hosting skill: npx skills add Agenta-AI/agenta-skills
2. Help me self-host Agenta with its repository.
For more details, read the self-hosting documentation.
Roadmap
Harnesses
- Claude Code
- Pi
- Codex
- Gemini
- OpenCode
- Create an issue to prioritize yours
Models
- OpenAI
- Anthropic
- OpenRouter
- Mistral AI
- Cohere
- Anyscale
- Perplexity AI
- DeepInfra
- Together AI
- Groq
- Google Gemini
- Azure
- AWS Bedrock
- MiniMax
- OpenAI-compatible models
- Self-hosted models (Ollama)
- Create an issue to prioritize yours
Agent runtimes
- Local runtime
- Daytona sandboxes
- Docker sandboxes
- E2B sandboxes
- AgentComputer
- Vercel
- Cloudflare
- Modal
- BoxLite
- Create an issue to prioritize yours
Features
- Schedules
- Events from connected apps
- MCP servers (API key + unauthenticated)
- Generic webhook triggers
- Additional MCP transports (OAuth)
- Channels (Slack, Telegram, Discord, Teams)
- Mobile version
See the complete roadmap. Want to help with one of these items? Open a discussion or contribute.
How Agenta compares
n8n, Activepieces, and Zapier
These products are designed for building workflows with predefined steps. Agenta is designed for work that requires an agent to plan, use tools, and adapt its approach. Use a workflow builder for predictable processes. Use Agenta when the work needs judgment, or when you want the same agent to work with you in chat and run in the background.
Claude Cowork
Claude Cowork provides a workspace built around Claude. Agenta is open source and lets you choose your harness and model, inspect the components of your agents, and run them interactively or in the background.
Claude Code, Codex, Pi, and OpenCode
These coding agents provide the execution layer that plans work and uses tools. Agenta adds the shared workspace around that execution layer: files, team access, triggers, versions, and traces. Agenta supports Claude Code, Pi, and Codex today. Support for more harnesses is on the roadmap.
Community and contributing
Agenta is open source under the MIT License. You can inspect the code, run it yourself, and help shape what we build next.
- Read the documentation
- Report a bug
- Request a feature or share an idea
- Read the contributing guide
- Join the Slack community
If Agenta is useful to you, star the repository and tell us what you build.
⭐ Star Agenta
Consider giving us a star! It helps us grow our community and gets Agenta in front of more developers.
Contributors ✨
Thanks goes to these wonderful people (emoji key):
This project follows the all-contributors specification. Contributions of any kind are welcome!