Agenta

v0.100.0 Feature

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

Published 14d 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

agents evaluation llm-as-a-judge llm-evaluation llm-framework llm-monitoring

+9 more

llm-observability llm-platform llm-playground llm-tools llmops observability prompt-engineering prompt-management rag-evaluation

Summary

AI summary

Enhancements to frontend message display, API data population for PydanticAI v2+ Logfire spans, and expanded entities testing.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Medium	Populate ag.data for PydanticAI v2+ Logfire spans in API. Populate ag.data for PydanticAI v2+ Logfire spans in API. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	Cleaned up meters, generalized scope/period, prepared global read limits. Cleaned up meters, generalized scope/period, prepared global read limits. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	Extended access controls and billing settings. Extended access controls and billing settings. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	Extended events beyond deployments. Extended events beyond deployments. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	Added fade+pill truncation for long chat messages in frontend. Added fade+pill truncation for long chat messages in frontend. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Feature	Medium	Rewrote BeautifiedJsonView as collapsible tree in frontend. Rewrote BeautifiedJsonView as collapsible tree in frontend. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Dependency	Medium	Updated model list in chore. Updated model list in chore. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Performance	Medium	Replaced reserved-input-key rejection with unwrap heuristic in SDK hotfix. Replaced reserved-input-key rejection with unwrap heuristic in SDK hotfix. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Bugfix	Medium	Resolved corrupted Docker files issue. Resolved corrupted Docker files issue. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Bugfix	Medium	Fixed annotation queue related issues. Fixed annotation queue related issues. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Refactor	Medium	Moved secrets and tools to monorepo packages (FE chore). Moved secrets and tools to monorepo packages (FE chore). Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Other	Medium	Extended test tooling in CI. Extended test tooling in CI. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—

Full changelog

What's Changed

docs(changelog): add annotation queues entry by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4348
[docs] Update API reference docs by @github-actions[bot] in https://github.com/Agenta-AI/agenta/pull/4350
[chore] Update model list by @junaway in https://github.com/Agenta-AI/agenta/pull/4349
[ci] Fix failing workflows by @junaway in https://github.com/Agenta-AI/agenta/pull/4352
[fix] Resolve corrupted docker files by @junaway in https://github.com/Agenta-AI/agenta/pull/4355
[fix] Clean up support attributes by @jp-agenta in https://github.com/Agenta-AI/agenta/pull/4325
[AGE-3763] fix(api): populate ag.data for PydanticAI v2+ Logfire spans by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4357
[Feat] Clean up meters, generalize scope/period, prepare global read limits by @junaway in https://github.com/Agenta-AI/agenta/pull/4360
feat(frontend): add fade+pill truncation for long chat messages by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4344
fix(frontend): rewrite BeautifiedJsonView as collapsible tree by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4343
[feat] Extend access controls and billing settings by @junaway in https://github.com/Agenta-AI/agenta/pull/4330
chore(entities): Layer 1 unit tests for @agenta/entities molecules by @bekossy in https://github.com/Agenta-AI/agenta/pull/4339
[feat] Extend events beyond deployments by @junaway in https://github.com/Agenta-AI/agenta/pull/4335
[FE chore] Move secrets and tools to monorepo packages by @ardaerzin in https://github.com/Agenta-AI/agenta/pull/4338
chore: layer 2 integration tests for @agenta/entities by @bekossy in https://github.com/Agenta-AI/agenta/pull/4365
Adding layer, fixing fern clients and api references by @junaway in https://github.com/Agenta-AI/agenta/pull/4367
fix: annotation queue related issues by @ashrafchowdury in https://github.com/Agenta-AI/agenta/pull/4333
hotfix(sdk): replace reserved-input-key rejection with unwrap heuristic by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/4362
[ci] Extend test tooling by @junaway in https://github.com/Agenta-AI/agenta/pull/4372

Full Changelog: https://github.com/Agenta-AI/agenta/compare/v0.99.9...v0.100.0

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