Agenta

v0.105.6 Feature

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

Published 7d 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

SDK now emits tool-output-denied, runner adds Claude client tools and extended‑thinking support, and web fixes the active‑tab locator.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Medium	Adds tool‑output‑denied emission so UI distinguishes denial from failure. Adds tool‑output‑denied emission so UI distinguishes denial from failure. Source: llm_adapter@2026-07-19 Confidence: high	—
Feature	Medium	Enables client tools on Claude in remote sandboxes. Enables client tools on Claude in remote sandboxes. Source: llm_adapter@2026-07-19 Confidence: high	—
Feature	Medium	Surfaces Claude extended‑thinking in the playground. Surfaces Claude extended‑thinking in the playground. Source: llm_adapter@2026-07-19 Confidence: high	—
Feature	Medium	Enables agent rename on the onboarding playground. Enables agent rename on the onboarding playground. Source: llm_adapter@2026-07-19 Confidence: high	—
Feature	Medium	Adds agent release gate — a portable wire‑level QA harness. Adds agent release gate — a portable wire‑level QA harness. Source: llm_adapter@2026-07-19 Confidence: high	—
Bugfix
Bugfix	Medium	Fixes SDK to use Claude harness model identifiers. Fixes SDK to use Claude harness model identifiers. Source: llm_adapter@2026-07-19 Confidence: high	—
Bugfix	Medium	Fixes active‑tab locator that broke five acceptance tests. Fixes active‑tab locator that broke five acceptance tests. Source: llm_adapter@2026-07-19 Confidence: high	—
Bugfix	Medium	Shows raw payload for commit deltas outside parameters in frontend. Shows raw payload for commit deltas outside parameters in frontend. Source: llm_adapter@2026-07-19 Confidence: high	—
Bugfix	Medium	Refreshes workflow name after rename. Refreshes workflow name after rename. Source: llm_adapter@2026-07-19 Confidence: high	—
Bugfix	Medium	Fixes typo exc_infp to exc_info in auth error logging. Fixes typo exc_infp to exc_info in auth error logging. Source: llm_adapter@2026-07-19 Confidence: low	—

Full changelog

What's Changed

test(web): fix the active-tab locator that broke five acceptance tests by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5378
feat(sdk): emit tool-output-denied so the UI can tell a denial from a failure by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5381
feat(runner): client tools on Claude in remote sandboxes by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5383
[AGE-3962] fix(runner): surface Claude extended-thinking in the playground by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5380
docs(readme): reposition around building and running agents by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5390
fix(auth): correct exc_infp typo to exc_info in auto-join error logging by @truecallerabreham in https://github.com/Agenta-AI/agenta/pull/5377
docs: add truecallerabreham as a contributor for bug by @allcontributors[bot] in https://github.com/Agenta-AI/agenta/pull/5379
fix(sdk): use Claude harness model identifiers by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5365
[fix] Refresh workflow name after rename by @ashrafchowdury in https://github.com/Agenta-AI/agenta/pull/5330
test(sdk): align bare Claude alias test with the curated alias list by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5393
test(agent): add the agent release gate — a portable wire-level QA harness by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/5351
[fix] Enable agent rename on the onboarding playground [AGE-3954] by @ashrafchowdury in https://github.com/Agenta-AI/agenta/pull/5340
fix(frontend): Show raw payload for commit deltas outside parameters by @ardaerzin in https://github.com/Agenta-AI/agenta/pull/5129
[release] v0.105.6 by @github-actions[bot] in https://github.com/Agenta-AI/agenta/pull/5385

New Contributors

@truecallerabreham made their first contribution in https://github.com/Agenta-AI/agenta/pull/5377

Full Changelog: https://github.com/Agenta-AI/agenta/compare/v0.105.5...v0.105.6

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

New Contributors

Related context

Related tools

Earlier breaking changes