agentaThe open-source workspace for building and running AI agents. Build agents through chat, share them with your team, and run background agents on schedules or app events.

agenta

Agenta is an open-source workspace for building, deploying, and managing AI agents that automate your work. Build agents through intuitive chat interfaces, collaborate with your team, and run background agents on schedules or app events—all while maintaining control over your data and using your existing AI subscriptions.

Key Features

• Chat-Based Agent Building: Create specialized agents by describing your workflow and connecting the apps they need, with continuous improvement through feedback loops
• Team Collaboration & Sharing: Share agents with team members, manage permissions by role, and work together in shared workspaces with documents and research organization
• Background Agent Automation: Schedule agents to run automatically or trigger them based on events from connected applications
• Comprehensive Tracing & Monitoring: Inspect every model and tool call, track token usage, maintain version history, and analyze costs for each agent
• Flexible Model & Harness Support: Switch between Claude, ChatGPT, and other LLMs without rebuilding agents; supports both API-based and self-hosted models

Use Cases

• Workflow Automation: Automate recurring tasks and business processes with approval gates and permission controls for security
• Team Productivity: Augment team capabilities by sharing specialized agents that handle research, document writing, and administrative work
• Cost-Efficient AI Operations: Self-host with your existing Claude or ChatGPT subscription instead of using metered API billing

Who Is It For

Teams and organizations looking to build reliable AI agents for internal automation without vendor lock-in. Ideal for companies wanting to self-host, maintain version control over agent configurations, and integrate AI capabilities with their existing tools through MCP servers or 1000+ app integrations.