HTTP 200 verified twice
Finding your next stop…
Loading the latest directory information.
Loading the latest directory information.
Start here for the decision-making essentials: what Agenta does, who it is for, how it is accessed, and the first-party sources behind this profile.
Hobby
Read-only public captures of Agenta’s homepage and verified pricing page. Screenshots are dated, never live embeds, and open full-screen.
Agenta is an open-source workspace for building, evaluating, and sharing AI agents through chat, with tracing, evaluation, and version control. Available as Agenta Cloud or self-hosted, it supports major LLM providers and includes human-in-the-loop controls and RBAC.
Short answers to the questions buyers and builders commonly ask about Agenta. Each answer cites the shared ledger below, where every source is listed once.
Build agents through chat, improve them with feedback, and share them with your whole team · Open-source edition includes chat, projects, agent configuration, instructions, skills, to · Open-source workspace for building and running agents · Build agents through chat, improve them with feedback, share them with your team
Build agents through chat, improve them with feedback, and share them with your whole team — self-hosted or in the cloud.
Build specialized agents for different areas of work and automate recurring tasks
Agenta lets you build specialized agents for different areas of work. You can work with them as AI coworkers, share them with your team, and use them to automate recurring tasks
Hobby | Free | 5,000 agent runs per monthly billing period, two team members · Pro and Business | 10,000 agent runs per month included · Additional agent runs | $5 per 10,000 runs (billed automatically, agents continue running) · Free tier available in Agenta Cloud
Hobby stops running agents after 5,000 agent runs in a monthly billing period.
Supports OpenAI, Anthropic, Gemini, Mistral, Groq, MiniMax, Together AI, OpenRouter, Azure · Agents built from standard components: AGENTS.md, skills, tools, and MCP servers
Connect OpenAI, Anthropic, Gemini, Mistral, Groq, MiniMax, Together AI, OpenRouter, Azure OpenAI, AWS Bedrock, or Google Vertex AI. You can also connect an OpenAI-compatible endpoint.
Self-hosted (open source, free) or Agenta Cloud (paid) · Self-hosted Agenta is deployed and operated in the customer's own infrastructure, with cus · Available as Agenta Cloud and as a self-hosted deployment · Available as Agenta Cloud and as an open-source self-hosted product
Self-host the open-source platform for free, or run on Agenta Cloud with prompt management, evaluation, and observability built in.
Trace data retention: Hobby 1 week, Pro 1 month, Business 3 months, Enterprise configurabl · Detailed permission system with human-in-the-loop controls for agent actions
Hobby retains trace data for one week, Pro for one month, and Business for three months. Enterprise retention is configurable.
This profile connects the jobs Agenta is described as handling with its delivery model, access options and the subjects used to match it to related products in this directory.
Each fact points to a recorded source, making it easy to distinguish verified product information from claims that need checking.
HTTP 200 verified twice
Agenta — The open-source workspace for your agents
Agenta is an open-source workspace for building and running agents · Agents can be built through chat, improved with feedback, and shared with a team · Available as Agenta Cloud and as a self-hosted product
Build agents through chat, improve them with feedback, and share them with your whole team
Yes — Agenta is an open-source workspace
Self-hosted (open source, free) or Agenta Cloud (paid)
Hobby | Free | 5,000 agent runs per monthly billing period, two team members
Pro and Business | 10,000 agent runs per month included
Additional agent runs | $5 per 10,000 runs (billed automatically, agents continue running)
Supports OpenAI, Anthropic, Gemini, Mistral, Groq, MiniMax, Together AI, OpenRouter, Azure OpenAI, AWS Bedrock, and Google Vertex AI; also OpenAI-compatible endpoints
Self-hosted Agenta is deployed and operated in the customer's own infrastructure, with customer-owned model and integration credentials
Trace data retention: Hobby 1 week, Pro 1 month, Business 3 months, Enterprise configurable
A concise view of the jobs, capabilities and integrations described in the recorded product sources.
Documented product formats, platforms and official distribution destinations. Availability can vary by region and plan.
A concise history of software releases and material product changes. Events appear only when a dated source supports what changed.
Showing the newest updates and meaningful milestones. Open an entry for its summary and source.
Improved sandbox security by replacing real API keys (model provider keys and HTTP MCP server credentials) in Daytona sandboxes with host-scoped placeholders, preventing agent exfiltration of credentials; on by default in Agenta Cloud.
View source [10]Agents can now run on OpenAI's Codex coding agent, with five OpenAI models, local or Daytona cloud sandboxes, existing tool approvals, HTTP MCP server support, and image attachments.
View source [3]Added support for running agents on OpenAI's Codex coding agent with five OpenAI models (GPT-5.6 Sol down to Luna), local or Daytona cloud sandbox execution, existing tool approvals, HTTP MCP servers, and image attachment reading.
View source [3]Agents can now be run on OpenAI's Codex coding harness across five GPT-5.6 Sol-to-Luna models, in local or Daytona cloud sandboxes, with HTTP MCP support and image attachment reading.
View source [3]Agents now operate from a shared cloud folder rather than files on one user's laptop, with persistent per-agent folders for reference material and cross-team file access.
View source [3]Added shared cloud-based workspace folders so uploaded images, PDFs, spreadsheets, and folders are visible to teammates and agents; each agent keeps a persistent folder across conversations, and Pi's seven built-in tools are always available.
View source [3]Released a rebuilt Playground that lets users attach evaluators to score outputs while editing and load test sets, combining prompt iteration with evaluation in one workbench.
View source [3]Added a dark theme across the application with light, dark, and system options accessible from the top bar; user choice persists between sessions.
View source [3]Facts, answers, structured details, milestones and primary resource links cite this shared ledger. Each external page appears once; release tags from the same GitHub project are grouped under one release history.
These tools share a workflow, capability or audience with Agenta. They may complement it rather than replace it, and are not presented as integrations or endorsements.
Open-source edition includes chat, projects, agent configuration, instructions, skills, tools, integrations, MCP servers, files, schedules, event triggers, tracing, evaluations, version history, role-based access control, and SSO
Enterprise tier adds audit logs, custom domains, deployment and upgrade support, priority support, private Slack Connect channel, dedicated support, and custom SLA
Agent package is AGENTS.md, skills, and MCP tools — harness, model, and runtime are swappable
Available as Agenta Cloud and as a self-hosted deployment
Web-based application accessible at https://cloud.agenta.ai
SOC 2 or other third-party audit reports available on request to demonstrate DPA compliance
Registered in Germany; data protection governed by the GDPR, the UK GDPR, and the Swiss Federal Act on Data Protection
Rudi-Dutschke-str. 23, Berlin 10969, Germany
Five OpenAI models available through the Codex harness, from GPT-5.6 Sol down to Luna, most with a context window of about one million tokens
Open-source workspace for building and running agents
Build agents through chat, improve them with feedback, share them with your team
Build specialized agents for different areas of work and automate recurring tasks
Yes — Agenta is an open-source, self-hosted product
Available as Agenta Cloud and as an open-source self-hosted product
Runs on Agenta Cloud, locally in Docker containers, or through Daytona sandboxes
Agents built from standard components: AGENTS.md, skills, tools, and MCP servers
Detailed permission system with human-in-the-loop controls for agent actions
Every agent run is logged and traced with inputs and outputs, and every agent configuration is versioned
Free tier available in Agenta Cloud
2026-07-22
Prompt management, evaluation, and observability for LLM apps and AI agents · Build agents through chat, improve them with feedback, and share them with the team · Agent package defined as AGENTS.md, skills, and MCP tools; harness, model, and runtime can
Prompt management, evaluation, and observability for LLM apps and AI agents
Build agents through chat, improve them with feedback, and share them with the team
Agent package defined as AGENTS.md, skills, and MCP tools; harness, model, and runtime can be swapped without code changes
Self-hosted or in the cloud (Agenta Cloud)
Open-source platform; the self-hosted edition is free
Additional agent runs are billed automatically at $5 per 10,000
Connects to model providers OpenAI, Anthropic, Gemini, Mistral, Groq, MiniMax, Together AI, OpenRouter, Azure OpenAI, AWS Bedrock, Google Vertex AI, plus any OpenAI-compatible endpoint
Added the ability to trigger automations on prompt deployments, sending events to any HTTPS endpoint or calling GitHub via repository_dispatch or workflow_dispatch to wire prompt deployments into CI, sync jobs, and PR workflows.
View source [3]
