AG

Agenta

Open-source LLMOps platform for prompt and app management

Self-Hosted AI ★ 4.8k stars Medium setup MIT

Agenta is an end-to-end LLMOps platform that combines prompt management, evaluation and observability for building production LLM applications. It can be fully self-hosted with Docker Compose.

Key features

  • Prompt playground
  • Evaluation suite
  • Observability tracing
  • Self-hosted with Docker

Strengths

  • Released under the MIT license
  • First-class Docker support for quick deployment
  • Active community (4.8k GitHub stars)
  • Written in TypeScript

Is Agenta secure?

No known vulnerabilities affect Agenta's latest release (v0.122.5). OSV.dev lists 2 past advisories for the project; none of them applies to this release. No known vulnerabilities is not proof of security: it means none have been publicly catalogued.

  • Most recent fixed advisory CVE-2026-27952, published 2026-02-26 (230 days ago), fixed in 0.48.1

Source: OSV.dev, checked 2026-10-14. How this is measured.

Agenta replaces

Health badge

Maintain Agenta? Add its health score to your README. The badge links back to this page.

SelfHostIndex: 92 · excellent

Markdown
[![SelfHostIndex health score for Agenta](https://selfhostindex.com/badge/agenta.svg)](https://selfhostindex.com/app/agenta/)
HTML
<a href="https://selfhostindex.com/app/agenta/"><img src="https://selfhostindex.com/badge/agenta.svg" alt="SelfHostIndex health score for Agenta"></a>
reStructuredText
.. image:: https://selfhostindex.com/badge/agenta.svg
   :alt: SelfHostIndex health score for Agenta
   :target: https://selfhostindex.com/app/agenta/

The score is recomputed on every nightly rebuild from public repository data — see how it is calculated.

Compare Agenta

8 head-to-head comparisons.

Similar self-hosted ai apps