Skip to content

openai-agents-js

v0.11.5 Feature

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

✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agents openai openai-api realtime-api typescript

ReleasePort's take

Light signal
editorial:auto 11d

Release v0.11.5 adds several new features and bug fixes across tracing, function tools, WebRTC handling, session persistence, runner configuration, usage JSON restoration, and AI SDK token preservation.

Why it matters: Provides opt‑in recovery for missing function tools, configurable tracing ID generation, scoped trace context helpers, delayed WebRTC close on transient disconnects, optional Runner modelProvider, and preserves AI SDK output token details in usage tracing—enabling more robust and customizable workflows.

Summary

AI summary

Updates feat, fix, and Documentation & Other Changes across a mixed release.

Changes in this release

Feature Medium

add opt-in recovery for missing function tools

add opt-in recovery for missing function tools

Source: llm_adapter@2026-05-23

Confidence: high

Feature Medium

add configurable tracing ID generation

add configurable tracing ID generation

Source: llm_adapter@2026-05-23

Confidence: high

Feature Medium

add Handoff clone overrides

add Handoff clone overrides

Source: llm_adapter@2026-05-23

Confidence: high

Feature Medium

add scoped trace context helpers

add scoped trace context helpers

Source: llm_adapter@2026-05-23

Confidence: high

Feature Medium

add RunState trace clearing for resumed runs

add RunState trace clearing for resumed runs

Source: llm_adapter@2026-05-23

Confidence: high

Feature Medium

add completed tracing lifecycle dispatch helpers

add completed tracing lifecycle dispatch helpers

Source: llm_adapter@2026-05-23

Confidence: high

Bugfix Medium

delay WebRTC close for transient peer disconnects

delay WebRTC close for transient peer disconnects

Source: llm_adapter@2026-05-23

Confidence: high

Bugfix Medium

add tracing context storage injection

add tracing context storage injection

Source: llm_adapter@2026-05-23

Confidence: high

Bugfix Medium

guard session persistence debug env access

guard session persistence debug env access

Source: llm_adapter@2026-05-23

Confidence: high

Bugfix Medium

make Runner modelProvider optional

make Runner modelProvider optional

Source: llm_adapter@2026-05-23

Confidence: high

Bugfix Medium

add public Usage JSON restoration helpers

add public Usage JSON restoration helpers

Source: llm_adapter@2026-05-23

Confidence: high

Bugfix Medium

preserve AI SDK output token details in usage tracing (issue #1340)

preserve AI SDK output token details in usage tracing (issue #1340)

Source: llm_adapter@2026-05-23

Confidence: high

Other Medium

add runtime review guidance

add runtime review guidance

Source: llm_adapter@2026-05-23

Confidence: low

Other Medium

update all translated pages using new settings

update all translated pages using new settings

Source: llm_adapter@2026-05-23

Confidence: low

Other Medium

clean up CI jobs

clean up CI jobs

Source: llm_adapter@2026-05-23

Confidence: low

Other Medium

cover MCP tools with disabled tracing in realtime tests

cover MCP tools with disabled tracing in realtime tests

Source: llm_adapter@2026-05-23

Confidence: low

Other Medium

update versions via CI bot

update versions via CI bot

Source: llm_adapter@2026-05-23

Confidence: low

Full changelog

What's Changed

  • feat: add opt-in recovery for missing function tools by @seratch in https://github.com/openai/openai-agents-js/pull/1336
  • feat: add configurable tracing ID generation by @seratch in https://github.com/openai/openai-agents-js/pull/1344
  • feat: add Handoff clone overrides by @seratch in https://github.com/openai/openai-agents-js/pull/1348
  • feat: add scoped trace context helpers by @seratch in https://github.com/openai/openai-agents-js/pull/1347
  • feat: add RunState trace clearing for resumed runs by @seratch in https://github.com/openai/openai-agents-js/pull/1356
  • feat: add completed tracing lifecycle dispatch helpers by @seratch in https://github.com/openai/openai-agents-js/pull/1355
  • fix: delay WebRTC close for transient peer disconnects by @seratch in https://github.com/openai/openai-agents-js/pull/1363
  • fix: add tracing context storage injection by @seratch in https://github.com/openai/openai-agents-js/pull/1345
  • fix: guard session persistence debug env access by @seratch in https://github.com/openai/openai-agents-js/pull/1354
  • fix: make Runner modelProvider optional by @seratch in https://github.com/openai/openai-agents-js/pull/1349
  • fix: add public Usage JSON restoration helpers by @seratch in https://github.com/openai/openai-agents-js/pull/1350
  • fix: #1340 preserve AI SDK output token details in usage tracing by @seratch in https://github.com/openai/openai-agents-js/pull/1342

Documentation & Other Changes

  • docs: add runtime review guidance by @seratch in https://github.com/openai/openai-agents-js/pull/1316
  • docs: update all translated pages using new settings by @seratch in https://github.com/openai/openai-agents-js/pull/1320
  • ci: clean up jobs by @seratch in https://github.com/openai/openai-agents-js/pull/1321
  • test(realtime): cover MCP tools with disabled tracing by @Herrtian in https://github.com/openai/openai-agents-js/pull/1338
  • chore: update versions by @github-actions[bot] in https://github.com/openai/openai-agents-js/pull/1339

New Contributors

  • @Herrtian made their first contribution in https://github.com/openai/openai-agents-js/pull/1338

Full Changelog: https://github.com/openai/openai-agents-js/compare/v0.11.4...v0.11.5

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 openai-agents-js

Get notified when new releases ship.

Sign up free

About openai-agents-js

A lightweight, powerful framework for multi-agent workflows and voice agents

All releases →

Beta — feedback welcome: [email protected]