Skip to content

Agenta

v0.103.5 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 1mo LLM Frameworks
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agent-builder agent-observability agent-orchestration agent-workspace agentic-ai agentic-workflow
+12 more
agents ai-agent ai-agents ai-automation ai-workflow-builder background-agents harness human-in-the-loop llm-evaluation llm-observability mcp self-hosted

Summary

AI summary

Fix redis queue overflow, improve frontend UI elements, and update readme and agents documentation.

Full changelog

What's Changed

  • [release] v0.103.4 by @github-actions[bot] in https://github.com/Agenta-AI/agenta/pull/4687
  • [fix] broken redis queues overflowing without maxlen by @junaway in https://github.com/Agenta-AI/agenta/pull/4691
  • doc(docs): move test set annotation queue from planned to in progress by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4679
  • doc(docs): mark annotation queues and in-playground evaluation as shipped on the roadmap by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4678
  • [feat][UX/UI] Hide mapping and preview until a test set is selected by @NamHT4Devlop in https://github.com/Agenta-AI/agenta/pull/4692
  • docs: add NamHT4Devlop as a contributor for code by @allcontributors[bot] in https://github.com/Agenta-AI/agenta/pull/4700
  • (readme) Add walkthrough to readme by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4698
  • fix(frontend): show correct breadcrumb label for SDK evals by @unfitcoder101 in https://github.com/Agenta-AI/agenta/pull/4619
  • fix(frontend): show user feedback as filterable fields in observability by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4652
  • [fix] Table header stays aligned during horizontal scroll in InfiniteVirtualTable by @NamHT4Devlop in https://github.com/Agenta-AI/agenta/pull/4697
  • fix(frontend): show correct variant name in commit modal for new vari… by @Sanket2329 in https://github.com/Agenta-AI/agenta/pull/4694
  • docs(agents): add entity display name rules for workflows by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4665
  • docs(docs): replace intro cover image with platform walkthrough video by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4699
  • fix(frontend): keep test set columns on locally-joined playground rows by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4701
  • [Feat]: improve cascade entity selector UI by @ashrafchowdury in https://github.com/Agenta-AI/agenta/pull/4630
  • [4535] fix(frontend): fix evaluator playground navigation from trace drawer by @axelray-dev in https://github.com/Agenta-AI/agenta/pull/4560
  • refactor(frontend): drop per-project fan-out from all batch fetchers by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4637
  • feat(examples): CV screening demo with feedback-to-deploy walkthrough by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4607
  • [fix] Resolve broken invites in OSS (again) by @junaway in https://github.com/Agenta-AI/agenta/pull/4704

New Contributors

  • @NamHT4Devlop made their first contribution in https://github.com/Agenta-AI/agenta/pull/4692
  • @unfitcoder101 made their first contribution in https://github.com/Agenta-AI/agenta/pull/4619

Full Changelog: https://github.com/Agenta-AI/agenta/compare/v0.103.4...v0.103.5

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track Agenta

Get notified when new releases ship.

Sign up free

About Agenta

LLMOps platform for prompt management, LLM evaluation, and observability. Build, evaluate, and monitor production-grade LLM applications with collaborative prompt engineering.

All releases →

Related context

Earlier breaking changes

  • v0.101.1 Renames `beautified` to `pretty`.
  • v0.100.7 Renames providers table Status column to API Key.

Beta — feedback welcome: [email protected]