1 Release published by 1 person

• v0.0.7

  published Mar 26, 2025

26 Pull requests merged by 13 people

• Add Langtrace to tracing.md

  #371 merged Mar 28, 2025

• Remove duplicate dependency definition of pynput

  #367 merged Mar 27, 2025

• Hotfix mcp example

  #365 merged Mar 27, 2025

• Expose the "store" parameter through ModelSettings

  #357 merged Mar 26, 2025

• Raise error on more invalid function schemas

  #356 merged Mar 26, 2025

• Tracing screenshot for MCP docs

  #355 merged Mar 26, 2025

• MCP example for SSE

  #354 merged Mar 26, 2025

• v0.0.7

  #353 merged Mar 26, 2025

• Mark handoff span as errored when multiple handoffs are requested

  #344 merged Mar 26, 2025

• Adding Git MCP server example

  #343 merged Mar 25, 2025

• [5/n] MCP tracing

  #342 merged Mar 25, 2025

• feat: Add Graphviz-based agent visualization functionality

  #147 merged Mar 25, 2025

• fix(examples): make sure audio playback finishes

  #340 merged Mar 25, 2025

• Make the reset behavior on tool use configurable

  #335 merged Mar 25, 2025

• [4/n] Add docs for MCP

  #338 merged Mar 25, 2025

• [3/n] Add an MCP stdio example

  #337 merged Mar 25, 2025

• [3/n] Add an MCP stdio example

  #322 merged Mar 25, 2025

• [4/n] Add docs for MCP

  #324 merged Mar 25, 2025

• [2/n] Add MCP support to Runner

  #321 merged Mar 25, 2025

• [1/n] Add MCP types to the SDK

  #320 merged Mar 25, 2025

• Fix parallel_tool_calls when False

  #333 merged Mar 25, 2025

• Fix potential infinite tool call loop by resetting tool_choice after …

  #263 merged Mar 25, 2025

• [0/n] Only run tests on py3.9, not mypy

  #319 merged Mar 25, 2025

• chore: ignore PyCharm .idea/ directory

  #317 merged Mar 24, 2025

• Fix Typos

  #258 merged Mar 24, 2025

• Improve tracing error messages

  #311 merged Mar 23, 2025

14 Pull requests opened by 14 people

• Support a callable model

  #312 opened Mar 24, 2025

• Add LiteLLM support in line with AsyncOpenAI and with reproducible examples

  #318 opened Mar 24, 2025

• Support Azure and Ollama APIs, let open for more APIs

  #330 opened Mar 25, 2025

• triage agent tool choice requried + respond to user

  #339 opened Mar 25, 2025

• Fact checking guardrails

  #347 opened Mar 26, 2025

• add Self type

  #361 opened Mar 27, 2025

• Update README.md

  #370 opened Mar 27, 2025

• The seed for the random number generator included

  #382 opened Mar 28, 2025

• Update tracing.md - Added Okahu-Monocle

  #385 opened Mar 28, 2025

• Fix incorrect slot name on `SpeechSpanData`

  #386 opened Mar 28, 2025

• Support Recursive Nodes in OpenAI Agents Draw Graph Visualization Extension

  #387 opened Mar 28, 2025

• Add reasoning parameter to ModelSettings

  #388 opened Mar 29, 2025

• Fix Docstring Reference in 'ToolsToFinalOutputFunction' in 'agent.py'

  #391 opened Mar 29, 2025

• Wait to run the Agent first step until Input guardrails is complete

  #394 opened Mar 29, 2025

25 Issues closed by 11 people

• No Traces Shown

  #31 closed Mar 30, 2025

• example @function_tool with optional parameters

  #43 closed Mar 30, 2025

• Support for Azure OpenAI APIs?

  #96 closed Mar 30, 2025

• Handoff to multiple agents and transfer back to triage agent

  #256 closed Mar 29, 2025

• MCP not support some open-source MCP servers

  #392 closed Mar 29, 2025

• Short term memory

  #374 closed Mar 28, 2025

• MCP tools are directly put into tool list, LLM cannot tell which server they come from

  #362 closed Mar 28, 2025

• autonomous agents

  #372 closed Mar 27, 2025

• Error if function_tool has a `Mapping` input

  #345 closed Mar 27, 2025

• Have any free model provider that can be used for structure output? (output_type argument when creating an agent)?

  #349 closed Mar 27, 2025

• Issue with incomplete pip installation - macOS

  #373 closed Mar 27, 2025

• How to get the Agent, using a LLM step in between tools, when needed?

  #364 closed Mar 27, 2025

• How to use the agent to process VQA data?

  #331 closed Mar 27, 2025

• OpenAIChatCompletionsModel is not worked

  #279 closed Mar 27, 2025

• Expose responses API `store` parameter

  #173 closed Mar 26, 2025

• The generated speech is slow, as if in slow-mo.

  #358 closed Mar 26, 2025

• Add MCP support

  #23 closed Mar 26, 2025

• How to provide object as input to agent

  #316 closed Mar 25, 2025

• tuple or namedtuple as a tool parameter type generates invalid schema

  #302 closed Mar 24, 2025

• Error processing output: BaseEventLoop.create_connection() got an unexpected keyword argument ‘additional_headers’

  #303 closed Mar 24, 2025

• 虽然开源了，，，但还是限制很多，太局限了，我没法扩展

  #62 closed Mar 24, 2025

• how to force use tool and send the result to llm？

  #281 closed Mar 24, 2025

• unsupported_country_region_territory

  #309 closed Mar 24, 2025

• error: the following arguments are required: -t/--tool-use-behavior

  #307 closed Mar 23, 2025

• OpenRouter as Custom Provider: "Invalid JSON" Error (Bad Request 400)

  #305 closed Mar 23, 2025

35 Issues opened by 31 people

• How to trans sdk output to chat completions

  #393 opened Mar 29, 2025

• How do we use `stream_events()` when `took_choice` = "function_name" ?

  #390 opened Mar 29, 2025

• Performance Issue with Agents - Synchronous API Call

  #389 opened Mar 29, 2025

• TTS starts after entire text is received

  #384 opened Mar 28, 2025

• Support For CodeAct In The Future?

  #383 opened Mar 28, 2025

• Rate limit in RunConfig

  #381 opened Mar 28, 2025

• Errors from custom model providers dont handled

  #380 opened Mar 28, 2025

• Availability on Azure

  #379 opened Mar 28, 2025

• human-in-the-loop

  #378 opened Mar 28, 2025

• How to implement tool calls and outputs flow between backend agents and client similar to Assistant API?

  #377 opened Mar 28, 2025

• Only use certain tools from MCP server

  #376 opened Mar 27, 2025

• The voice models aren't waiting for a pause(a very natural phenomenon when using a voice agent) and generating an output mid sentence when talking to the voice agent implemented using "VoicePipline" .

  #369 opened Mar 27, 2025

• Random transcript gets printed/generated when talking to the voice agent implemented using "VoicePipline" . Eg - "Transcription: Kurs." Mind you there is no background noise.

  #368 opened Mar 27, 2025

• Implementing a comprehensive workflow

  #366 opened Mar 27, 2025

• windows run mcp error：

  #363 opened Mar 27, 2025

• how to use Code Interpreter or Image Output in OpenAI Agents SDK

  #360 opened Mar 27, 2025

• The depth agents handsoff draw_graph function can auto find out?

  #359 opened Mar 27, 2025

• Mermaid-based visualization

  #352 opened Mar 26, 2025

• how to get the result of model output

  #351 opened Mar 26, 2025

• Golang implementation / alternative

  #350 opened Mar 26, 2025

• context management between the agents in a multi agent set up

  #348 opened Mar 26, 2025

• Access to Chat History in Hooks

  #346 opened Mar 26, 2025

• Add Support for Image Return in Agent Tools

  #341 opened Mar 25, 2025

• How to process the data of the reasoning model？

  #334 opened Mar 25, 2025

• Custom model provider agent does not use tool calling when an output type is specified

  #332 opened Mar 25, 2025

• Feature Request: Support Manual Interruption of OpenAI Agent During Execution

  #329 opened Mar 25, 2025

• Returning function call's responses in `raw_response_event`

  #328 opened Mar 25, 2025

• `tts_settings` should be available per Agent

  #327 opened Mar 25, 2025

• Question about Tool Call Streaming

  #326 opened Mar 25, 2025

• Retry mechanism for ModelBehaviorError

  #325 opened Mar 25, 2025

• Get cost for an agent

  #323 opened Mar 24, 2025

• UTF-8 encoding issues in tool calling params with GPT-4o model

  #315 opened Mar 24, 2025

• Duplicate declarations of variables in source code

  #314 opened Mar 24, 2025

• Can I use multiple Agents as a single Agent?

  #313 opened Mar 24, 2025

• Callable model

  #310 opened Mar 23, 2025

22 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Add tool call parameters for `on_tool_start` hook

  #253 commented on Mar 26, 2025 • 1 new comment

• Add FastAPI example

  #168 commented on Mar 27, 2025 • 1 new comment

• Create Agent solathan file

  #308 commented on Mar 23, 2025 • 0 new comments

• chore: add missing openai.fm model names to `TTSModelSettings.voice` type

  #269 commented on Mar 29, 2025 • 0 new comments

• Add reasoning_effort parameter to ModelSettings Fixes #189

  #235 commented on Mar 23, 2025 • 0 new comments

• test: extended test suite with additional tests

  #91 commented on Mar 24, 2025 • 0 new comments

• perf: implement screenshot caching via Responses API reference IDs

  #90 commented on Mar 26, 2025 • 0 new comments

• Add Previous/Next Navigation to MkDocs Footer

  #79 commented on Mar 24, 2025 • 0 new comments

• OPENAI_API_KEY is not set, skipping trace export

  #190 commented on Mar 29, 2025 • 0 new comments

• what is the best practices to have faster voices conversation?

  #306 commented on Mar 28, 2025 • 0 new comments

• Enhance `on_tool_start` Hook to Include Tool Call Arguments

  #252 commented on Mar 28, 2025 • 0 new comments

• invalid_request_error when using "chat_completions" with triage agent (gemini -> any other model)

  #237 commented on Mar 27, 2025 • 0 new comments

• Integrate more 3rd party tools into AgentSDK tool

  #299 commented on Mar 27, 2025 • 0 new comments

• Issue when processing real time audio from a Twilio media stream

  #304 commented on Mar 25, 2025 • 0 new comments

• How to service agents over HTTP

  #126 commented on Mar 25, 2025 • 0 new comments

• Return

  #259 commented on Mar 25, 2025 • 0 new comments

• Streamed Voice Agent Demo - Multiple Performance Issues

  #301 commented on Mar 25, 2025 • 0 new comments

• Handoff does not work with Claude 3.7 Sonnet

  #270 commented on Mar 25, 2025 • 0 new comments

• Support for OpenAI agents sdk with Javascript/Typescript

  #240 commented on Mar 24, 2025 • 0 new comments

• Support AWS Bedrock

  #86 commented on Mar 24, 2025 • 0 new comments

• Support for OpenAI Realtime Models in Agents SDK

  #239 commented on Mar 23, 2025 • 0 new comments

• While streaming input guardrail triggers after the streaming of actual answer starts

  #300 commented on Mar 23, 2025 • 0 new comments