agenta.ai

mx

smtp.google.com

ns

hassan.ns.cloudflare.com

pola.ns.cloudflare.com

caa

txt

google-site-verification=hUm84OcvseG6fwGZ-FdruRDRq96dh-RqYY1GSMJmfFE

v=spf1 include:_spf.google.com ~all

google-site-verification=dYPtXbTPppSm8X0PpKq3dZn8XrM6eoc_nXjAn_0W1fw

google-site-verification=MKZIX__XbIoeza9Nb-xhIs1Y9nCg0zT_1rfK3nOg5l8

dmarc

v=DMARC1; p=none; rua=mailto:2bd591694ec94f4084365d10c36c7582@dmarc-reports.cloudflare.net

# Agenta marketing site (https://agenta.ai)
# Allow all crawlers full access; point them at the generated sitemap index.
# The docs sitemap is listed too because agenta.ai/docs/* proxies the docs site.
User-agent: *
Allow: /

Sitemap: https://agenta.ai/sitemap-index.xml
Sitemap: https://agenta.ai/docs/sitemap.xml

# Agenta

> Agenta is the open-source workspace for your agents: build agents through chat,
> improve them with feedback, and share them with your whole team. Open source and
> self-hostable, or hosted in the cloud.

## About

- Build agents through chat: describe the job, give the agent context and tools, and improve it through real work and feedback.
- Share agents with your team, and run them on a schedule or when an event happens in a connected app.
- Consequential actions can wait for human approval before they run.
- Prompts, skills, and tools are versioned like code, so you can roll back to any revision.
- MIT-licensed and yours to run: self-host on your own infrastructure to keep your agents and data with you.

## When to use Agenta

- Build an agent from a plain-language description, then improve it with feedback from real runs.
- Version prompts, skills, and tools like code, and roll back to any earlier revision.
- Evaluate a change against a test set before it ships, with LLM-as-a-judge or your own evaluators.
- Trace every run to see each step, its cost, and where it failed.
- Run an agent on a schedule, or when an event happens in a connected app.
- Put a human approval step in front of consequential actions.

## When not to use Agenta

- A single model call with no versioning, evaluation, or tracing: call the model provider directly.
- Hosting or serving a model. Agenta orchestrates and observes agents; it is not an inference provider.

## How to call Agenta

- REST API: https://us.cloud.agenta.ai/api (US) or https://eu.cloud.agenta.ai/api (EU). Self-hosted: $AGENTA_HOST/api.
- Authenticate with `Authorization: ApiKey $AGENTA_API_KEY`. A key is scoped to one project, so no project_id parameter is needed.
- Full machine-readable surface: https://agenta.ai/openapi.json (OpenAPI 3.1).
- Python SDK: `pip install agenta`.
- Endpoint reference and guides: https://agenta.ai/api and https://docs.agenta.ai.

## Links

- [Website](https://agenta.ai): the marketing site.
- [Documentation](https://docs.agenta.ai): guides and API reference.
- [GitHub](https://github.com/Agenta-AI/agenta): the open-source repository.
- [Pricing](https://agenta.ai/pricing): hosted plans and the open-source tier.
- [Blog](https://agenta.ai/blog): articles and updates.

## Hosting

- Cloud: hosted at https://cloud.agenta.ai — start free, no infrastructure to run.
- Self-hosted: run Agenta on your own infrastructure under the MIT license.

## API

- OpenAPI specification: https://agenta.ai/openapi.json
- US base URL: https://us.cloud.agenta.ai/api
- EU base URL: https://eu.cloud.agenta.ai/api

## Machine-readable

- Sitemap: https://agenta.ai/sitemap-index.xml
- Every page also serves a markdown representation: request it with `Accept: text/markdown`, or append `.md` to the path (the homepage twin is `/index.md`).