Agenta

v0.103.3 Feature

This release adds 2 notable features for engineering teams evaluating rollout.

Published 1mo LLM Frameworks

View tool

✓ 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

Add bulk‑archive actions, self‑serve account deletion, and multiple frontend bugfixes.

Changes in this release

Type	Severity	Summary	CVE
Feature	Medium	Add bulk‑archive actions for apps, testsets and evaluators. Add bulk‑archive actions for apps, testsets and evaluators. Source: llm_adapter@2026-06-12 Confidence: high	—
Feature	Medium	Enable self-serve account deletion from Settings. Enable self-serve account deletion from Settings. Source: granite4.1:30b@2026-06-12-audit Confidence: low	—
Bugfix
Bugfix	Medium	Clear playground draft test cases when switching applications. Clear playground draft test cases when switching applications. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Preserve connected testset when opening Manage testcases. Preserve connected testset when opening Manage testcases. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Resolve broken `scenario` `status` in `annotation` `queues`. Resolve broken `scenario` `status` in `annotation` `queues`. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Keep test set synced across playground commits. Keep test set synced across playground commits. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Show boolean evaluator metrics as True/False, not a mean. Show boolean evaluator metrics as True/False, not a mean. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Default variant name to its slug when not provided. Default variant name to its slug when not provided. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Resolve broken `references` in Playground. Resolve broken `references` in Playground. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Keep unused test set columns on playground rows through Run. Keep unused test set columns on playground rows through Run. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Medium	Resolve broken filtering for schemaless `evaluators`. Resolve broken filtering for schemaless `evaluators`. Source: llm_adapter@2026-06-12 Confidence: high	—
Bugfix	Low	Label evaluator nodes in the playground by their entity name. Label evaluator nodes in the playground by their entity name. Source: granite4.1:30b@2026-06-12-audit Confidence: low	—
Bugfix	Low	Resolve evaluation names for SDK‑created apps and new evaluators. Resolve evaluation names for SDK‑created apps and new evaluators. Source: granite4.1:30b@2026-06-12-audit Confidence: low	—
Refactor	Low	Refactor workflow name selectors. Refactor workflow name selectors. Source: granite4.1:30b@2026-06-12-audit Confidence: low	—

Full changelog

What's Changed

[release] v0.103.2 by @github-actions[bot] in https://github.com/Agenta-AI/agenta/pull/4631
docs: add unfitcoder101 as a contributor for bug by @allcontributors[bot] in https://github.com/Agenta-AI/agenta/pull/4651
fix(frontend): clear playground draft test cases when switching applications by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4628
[Fix]: Preserve connected testset when opening Manage testcases by @ashrafchowdury in https://github.com/Agenta-AI/agenta/pull/4602
[fix] Resolve broken scenario status in annotation queues by @junaway in https://github.com/Agenta-AI/agenta/pull/4644
fix(frontend): keep the test set synced across playground commits by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4653
fix(frontend): show boolean evaluator metrics as True/False, not a mean by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4667
fix(api): default variant name to its slug when not provided by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4664
[fix] Resolve broken references in Playground by @junaway in https://github.com/Agenta-AI/agenta/pull/4656
[Feat]: Add bulk-archive actions for apps, testsets and evaluators by @ashrafchowdury in https://github.com/Agenta-AI/agenta/pull/4629
[fix] Keep unused test set columns on playground rows through Run by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4649
[fix] Resolve broken filtering for schemaless evaluators by @junaway in https://github.com/Agenta-AI/agenta/pull/4650
feat(api): Self-serve account deletion from Settings by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4600
fix(frontend): label evaluator nodes in the playground by their entity name by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4668
fix(frontend): resolve evaluation names for SDK-created apps and new evaluators by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4661
Refactor/workflow name selectors by @bekossy in https://github.com/Agenta-AI/agenta/pull/4685
[FE / Feat] Add evaluators to existing evals by @ardaerzin in https://github.com/Agenta-AI/agenta/pull/4577

Full Changelog: https://github.com/Agenta-AI/agenta/compare/v0.103.2...v0.103.3

View diff on GitHub

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

Share on X Share on Bluesky

Track Agenta

Get notified when new releases ship.

About Agenta

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

Agenta

Summary

Changes in this release

What's Changed

Related context

Related tools

Earlier breaking changes