# \#python **URL:** https://community.openai.com/tag/python/45.md [Latest](https://community.openai.com/latest.md) · [Categories](https://community.openai.com/categories.md) · [Tags](https://community.openai.com/tags.md) --- ## [How should enterprise AI agents cache, refresh, and select skills in runtime?](https://community.openai.com/t/how-should-enterprise-ai-agents-cache-refresh-and-select-skills-in-runtime/1406914)
**Author:** [@suresandeep03](https://community.openai.com/u/suresandeep03)\ **Replies:** 4\ **Last updated:** [October 17, 2026, 6:45am UTC](https://community.openai.com/t/how-should-enterprise-ai-agents-cache-refresh-and-select-skills-in-runtime/1406914 "2026-10-17T06:45:15Z")
Hi there, I’m are building an AI Agent that uses skill.md files. Each skill.md contains: Skill metadata skill\_name description examples business domain tags Execution instructions SQL query itself Snowflake execut… --- ## [Custom GPT not able to execute skill's script](https://community.openai.com/t/custom-gpt-not-able-to-execute-skills-script/1390915)
**Author:** [@AlexKras](https://community.openai.com/u/AlexKras)\ **Replies:** 1\ **Last updated:** [October 9, 2026, 6:20pm UTC](https://community.openai.com/t/custom-gpt-not-able-to-execute-skills-script/1390915 "2026-10-09T18:20:07Z")
I am creating a custom GPT to be a Business Assistant. When the user asks it to perform a certain action, it finds an associated skill and follows the instructions in the skill. The skill generates a report and has pytho… --- ## [ChatGPT 5.6 uses the offline Python/shell sandbox to do HTTP requests which always fail](https://community.openai.com/t/chatgpt-5-6-uses-the-offline-python-shell-sandbox-to-do-http-requests-which-always-fail/1386960)
**Author:** [@naive\_that5z](https://community.openai.com/u/naive_that5z)\ **Replies:** 4\ **Last updated:** [September 27, 2026, 12:28am UTC](https://community.openai.com/t/chatgpt-5-6-uses-the-offline-python-shell-sandbox-to-do-http-requests-which-always-fail/1386960 "2026-09-27T00:28:25Z")
Summary In standalone ChatGPT Chat, GPT-5.6 Sol Thinking in both Extra High and Pro modes repeatedly attempts to retrieve external web resources through Python, requests, curl, or shell commands executed inside the offli… --- ## [Using Codex to generate Blender Python scripts for Roblox creatures](https://community.openai.com/t/using-codex-to-generate-blender-python-scripts-for-roblox-creatures/1400919)
**Author:** [@bahous](https://community.openai.com/u/bahous)\ **Replies:** 0\ **Last updated:** [September 26, 2026, 7:53am UTC](https://community.openai.com/t/using-codex-to-generate-blender-python-scripts-for-roblox-creatures/1400919 "2026-09-26T07:53:39Z")
Hi, I am developing an original Roblox horror game and I create my creatures in Blender 5.2.1 LTS. I want to know which OpenAI/Codex workflow would be best for generating and fixing Blender Python (bpy) scripts for my … --- ## [Codex shows no diffs when files edited by python script](https://community.openai.com/t/codex-shows-no-diffs-when-files-edited-by-python-script/1395244)
**Author:** [@Mattzq](https://community.openai.com/u/Mattzq)\ **Replies:** 2\ **Last updated:** [September 7, 2026, 1:01pm UTC](https://community.openai.com/t/codex-shows-no-diffs-when-files-edited-by-python-script/1395244 "2026-09-07T13:01:42Z")
v. 0.153.3 Since Astra (medium) I catch agents responding that they changed files and yet I see no diffs. Turns out that’s because they use python scripts in bash to edit files. I assume that apply\_patch passes changes… --- ## [OpenAI gpt-5-mini with\_structured\_output returns malformed JSON filled with whitespace instead of valid response](https://community.openai.com/t/openai-gpt-5-mini-with-structured-output-returns-malformed-json-filled-with-whitespace-instead-of-valid-response/1377686)
**Author:** [@Gevorg\_Gevorgyan](https://community.openai.com/u/Gevorg_Gevorgyan)\ **Replies:** 1\ **Last updated:** [August 24, 2026, 4:34pm UTC](https://community.openai.com/t/openai-gpt-5-mini-with-structured-output-returns-malformed-json-filled-with-whitespace-instead-of-valid-response/1377686 "2026-08-24T16:34:36Z")
I’m using LangChain’s ChatOpenAI.with\_structured\_output() with gpt-5-mini to classify user intent into a Pydantic model. Intermittently, the model returns malformed JSON — the response starts with a valid key but never c… --- ## [My Python input utility for console chatbots - multiline input method, with editing and mid-prompt commands](https://community.openai.com/t/my-python-input-utility-for-console-chatbots-multiline-input-method-with-editing-and-mid-prompt-commands/1385734)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 1\ **Last updated:** [July 4, 2026, 9:43pm UTC](https://community.openai.com/t/my-python-input-utility-for-console-chatbots-multiline-input-method-with-editing-and-mid-prompt-commands/1385734 "2026-07-04T21:43:09Z")
This is a utility method that I’ve refined and used for hundreds of API calls - a broad replacement for input() for about every AI prompting scenario: where you want to type naturally within a Python console-based applic… --- ## [Codex app-server python SDK (experimental)](https://community.openai.com/t/codex-app-server-python-sdk-experimental/1376980)
**Author:** [@aprendendo.next](https://community.openai.com/u/aprendendo.next)\ **Replies:** 1\ **Last updated:** [May 16, 2026, 12:34pm UTC](https://community.openai.com/t/codex-app-server-python-sdk-experimental/1376980 "2026-05-16T12:34:45Z")
It seems a new codex SDK for python is being cooked: Mentioned in Codex 0.115.0 How to experiment with it: --- ## [Allow \`"annotation-only" keywords (like \`description\`, \`title\`, and \`examples\`) for $ref keywords when using JSON structured output](https://community.openai.com/t/allow-annotation-only-keywords-like-description-title-and-examples-for-ref-keywords-when-using-json-structured-output/1374904)
**Author:** [@colin10](https://community.openai.com/u/colin10)\ **Replies:** 1\ **Last updated:** [February 23, 2026, 8:45pm UTC](https://community.openai.com/t/allow-annotation-only-keywords-like-description-title-and-examples-for-ref-keywords-when-using-json-structured-output/1374904 "2026-02-23T20:45:36Z")
When using structured output via the API, the JSON validator rejects schemas where a $ref keyword has extra sibling keywords, like description: { type: object, properties: { foo: { $ref: #/$defs/Foo, … --- ## [OpenAI API and structured logging in Python](https://community.openai.com/t/openai-api-and-structured-logging-in-python/1374262)
**Author:** [@tonyaldon](https://community.openai.com/u/tonyaldon)\ **Replies:** 0\ **Last updated:** [February 15, 2026, 6:32am UTC](https://community.openai.com/t/openai-api-and-structured-logging-in-python/1374262 "2026-02-15T06:32:45Z")
Hi, I wrote a blog post with a reusable Python snippet that logs every OpenAI API request and response in JSON, including API errors. You might find it interesting :wink: tl;dr: I like the OpenAI dashboard Logs, but AP… --- ## [Python packages listing of what is available in API code interpreter](https://community.openai.com/t/python-packages-listing-of-what-is-available-in-api-code-interpreter/1365812)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 1\ **Last updated:** [November 10, 2025, 5:40pm UTC](https://community.openai.com/t/python-packages-listing-of-what-is-available-in-api-code-interpreter/1365812 "2025-11-10T17:40:22Z")
You might find it useful to provide an AI application that you know will be employing code interpreter the listing of Python modules that are installed in its notebook environment. You can list them in a system (develope… --- ## [Building a WhatsApp Travel Concierge with GPT-4o + Whisper + Flask](https://community.openai.com/t/building-a-whatsapp-travel-concierge-with-gpt-4o-whisper-flask/1363780)
**Author:** [@michelone](https://community.openai.com/u/michelone)\ **Replies:** 0\ **Last updated:** [October 24, 2025, 2:51pm UTC](https://community.openai.com/t/building-a-whatsapp-travel-concierge-with-gpt-4o-whisper-flask/1363780 "2025-10-24T14:51:49Z")
Hey everyone, I wanted to share a project I’ve been developing called Chat with Bella — a real-time travel assistant that runs entirely on WhatsApp. It uses: GPT-4o for conversational responses (multi-language, cont… --- ## [Introducing OpenAI HTTP proxy](https://community.openai.com/t/introducing-openai-http-proxy/1362633)
**Author:** [@Nayjest](https://community.openai.com/u/Nayjest)\ **Replies:** 0\ **Last updated:** [October 15, 2025, 5:10pm UTC](https://community.openai.com/t/introducing-openai-http-proxy/1362633 "2025-10-15T17:10:18Z")
Hello everyone, I’m pleased to announce the stable release of LM-Proxy — an OpenAI-compatible opensource HTTP proxy for multi-provider LLM inference (Google, Anthropic, OpenAI, PyTorch). Built with Python, FastAPI and … --- ## [Python + React : how to use ChatKit on localhost?](https://community.openai.com/t/python-react-how-to-use-chatkit-on-localhost/1361785)
**Author:** [@Jonathan\_Filocco](https://community.openai.com/u/Jonathan_Filocco)\ **Replies:** 0\ **Last updated:** [October 9, 2025, 9:10pm UTC](https://community.openai.com/t/python-react-how-to-use-chatkit-on-localhost/1361785 "2025-10-09T21:10:59Z")
Hello, I’m working on a frontend React App with Vite and a backend python server with FastAPI. I want to use ChatKit to improve my app chatbot. However, I’m not able to run it correctly as localhost is not a verified … --- ## [Tasks: Code interpreter (Python) not available - bug or intended?](https://community.openai.com/t/tasks-code-interpreter-python-not-available-bug-or-intended/1358807)
**Author:** [@jezek](https://community.openai.com/u/jezek)\ **Replies:** 2\ **Last updated:** [September 21, 2025, 4:48pm UTC](https://community.openai.com/t/tasks-code-interpreter-python-not-available-bug-or-intended/1358807 "2025-09-21T16:48:57Z")
In chat, GPT-5 can use the Python tool to make e.g. charts. The same instruction, when scheduled as a Task, won’t run Python and reports it isn’t available in the Task environment. Web browsing and memory still work. Is… --- ## [Keep Getting "You exceeded your current quota, please check your plan and billing details." in Python](https://community.openai.com/t/keep-getting-you-exceeded-your-current-quota-please-check-your-plan-and-billing-details-in-python/662017)
**Author:** [@fanmixco](https://community.openai.com/u/fanmixco)\ **Replies:** 6\ **Last updated:** [September 9, 2025, 9:05pm UTC](https://community.openai.com/t/keep-getting-you-exceeded-your-current-quota-please-check-your-plan-and-billing-details-in-python/662017 "2025-09-09T21:05:52Z")
Fellows, I’m not sure if I’m the only one, but I keep getting: “You exceeded your current quota, please check your plan and billing details. For more information on this error.” My app is extremely basic: from langc… --- ## [Python code not able to run in Canvas](https://community.openai.com/t/python-code-not-able-to-run-in-canvas/1080127)
**Author:** [@jordi27](https://community.openai.com/u/jordi27)\ **Replies:** 8\ **Last updated:** [July 27, 2025, 4:10am UTC](https://community.openai.com/t/python-code-not-able-to-run-in-canvas/1080127 "2025-07-27T04:10:35Z")
Hi, I’ve ran into the problem that running python code in Canvas is no longer possible. I have ChatGPT Plus. It has always worked for me, is this due to the fact that I’m at my limit or is this a bug? As you can see in… --- ## [How can I get the progress percentage of image generation with the OpenAI Image API?](https://community.openai.com/t/how-can-i-get-the-progress-percentage-of-image-generation-with-the-openai-image-api/1303151)
**Author:** [@Muhammed\_Emin\_Kose](https://community.openai.com/u/Muhammed_Emin_Kose)\ **Replies:** 1\ **Last updated:** [June 30, 2025, 12:52pm UTC](https://community.openai.com/t/how-can-i-get-the-progress-percentage-of-image-generation-with-the-openai-image-api/1303151 "2025-06-30T12:52:01Z")
Is there a way to retrieve real‑time progress (e.g. 0%–100%) when generating images via the OpenAI Image API? --- ## [Token Aware semantic based chunker - https://pypi.org/project/semantic-chunker-langchain/](https://community.openai.com/t/token-aware-semantic-based-chunker-https-pypi-org-project-semantic-chunker-langchain/1299116)
**Author:** [@Prajwal\_Mandale333](https://community.openai.com/u/Prajwal_Mandale333)\ **Replies:** 0\ **Last updated:** [June 26, 2025, 6:37am UTC](https://community.openai.com/t/token-aware-semantic-based-chunker-https-pypi-org-project-semantic-chunker-langchain/1299116 "2025-06-26T06:37:53Z")
pip install semantic-chunker-langchain There is an issue with the models handling the large size of the pdf , txt files etc and when we pass the models like gpt 3.5 cant handle the 1000 + pages pdf result in token limit… --- ## [Batch file documentation error - and many more documentation issues added](https://community.openai.com/t/batch-file-documentation-error-and-many-more-documentation-issues-added/1246138)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 20\ **Last updated:** [June 5, 2025, 6:54pm UTC](https://community.openai.com/t/batch-file-documentation-error-and-many-more-documentation-issues-added/1246138 "2025-06-05T18:54:24Z")
You will find that a Python execution environment doesn’t appreciate JavaScript’s “const” very much… What I need to find by doing (and then waiting 24h) instead of by reading anywhere it is offered: the files endpoi… --- ## [Why do I keep getting AceConnectionException?](https://community.openai.com/t/why-do-i-keep-getting-aceconnectionexception/1279148)
**Author:** [@EdgeCrafter](https://community.openai.com/u/EdgeCrafter)\ **Replies:** 0\ **Last updated:** [June 5, 2025, 1:48pm UTC](https://community.openai.com/t/why-do-i-keep-getting-aceconnectionexception/1279148 "2025-06-05T13:48:47Z")
I keep getting AceConnectionException Exploring root cause ChatGPT reports Python rendering kernel remains unstable New session or giving a reboot sandbox command does not help :frowning: Any others with the same iss… --- ## [Codex: Python dependencies install, but agent can't see them](https://community.openai.com/t/codex-python-dependencies-install-but-agent-cant-see-them/1268801)
**Author:** [@dbb](https://community.openai.com/u/dbb)\ **Replies:** 4\ **Last updated:** [June 4, 2025, 7:02am UTC](https://community.openai.com/t/codex-python-dependencies-install-but-agent-cant-see-them/1268801 "2025-06-04T07:02:13Z")
Codex is unable to see Python dependencies. I tried three experiments in order to rule out user error. This can be reproduced by doing the following: Create a repository containing a requirements.txt with one dependenc… --- ## [AI development course #beginner A1](https://community.openai.com/t/ai-development-course-beginner-a1/1138722)
**Author:** [@jochenschultz](https://community.openai.com/u/jochenschultz)\ **Replies:** 32\ **Last updated:** [June 3, 2025, 7:44pm UTC](https://community.openai.com/t/ai-development-course-beginner-a1/1138722 "2025-06-03T19:44:57Z")
When people in this forum discuss AI ethics, consciousness, self-awareness, and related topics, I often notice a lack of basic technical understanding. So, I thought: Why not start a beginner-friendly course right here? … --- ## [Problem with Responses API + Stream + Function Calling with Openai Python SDK](https://community.openai.com/t/problem-with-responses-api-stream-function-calling-with-openai-python-sdk/1266970)
**Author:** [@ciro.artigot](https://community.openai.com/u/ciro.artigot)\ **Replies:** 8\ **Last updated:** [May 22, 2025, 7:51pm UTC](https://community.openai.com/t/problem-with-responses-api-stream-function-calling-with-openai-python-sdk/1266970 "2025-05-22T19:51:45Z")
Hello, I have a problem ini my python implementation with openai SDK using the responses API with stream and function calling. It´s all ok, the endpoint works fine streaming text messages and file search. I’m implemen… --- ## [Python\> Termux Error (Novice - Learning)](https://community.openai.com/t/python-termux-error-novice-learning/1252489)
**Author:** [@redknight\_I](https://community.openai.com/u/redknight_I)\ **Replies:** 3\ **Last updated:** [May 5, 2025, 11:23pm UTC](https://community.openai.com/t/python-termux-error-novice-learning/1252489 "2025-05-05T23:23:55Z")
from openai import OpenAI client = OpenAI() response = client.responses.create( model=“gpt-4.1”, input=“Write a one-sentence bedtime story about a unicorn.” ) print(response.output\_text) Error: from openai import… --- ## [Invite? Possibly? Help and advice! Baby developer with training wheels!](https://community.openai.com/t/invite-possibly-help-and-advice-baby-developer-with-training-wheels/1240437)
**Author:** [@JaK0909](https://community.openai.com/u/JaK0909)\ **Replies:** 1\ **Last updated:** [April 24, 2025, 10:29am UTC](https://community.openai.com/t/invite-possibly-help-and-advice-baby-developer-with-training-wheels/1240437 "2025-04-24T10:29:54Z")
I’m very new to this and through the love of ai I have developed a massive interest in developing - currently working with the bots to make something I think is very cool but will be basic to you guys I imagine. I’ve do… --- ## [Pydantic integration seems to be broken in new responses.create API](https://community.openai.com/t/pydantic-integration-seems-to-be-broken-in-new-responses-create-api/1230780)
**Author:** [@yuriy.ward](https://community.openai.com/u/yuriy.ward)\ **Replies:** 2\ **Last updated:** [April 17, 2025, 11:28am UTC](https://community.openai.com/t/pydantic-integration-seems-to-be-broken-in-new-responses-create-api/1230780 "2025-04-17T11:28:35Z")
Hey! I was previously able to use Pydantic types and field descriptions to define a schema with the Completions API – it worked great! Here’s an example that used to work: from openai import OpenAI from pydantic import… --- ## [\[Realtime API\] Error committing input audio buffer: the buffer is empty](https://community.openai.com/t/realtime-api-error-committing-input-audio-buffer-the-buffer-is-empty/975381)
**Author:** [@emkataumre](https://community.openai.com/u/emkataumre)\ **Replies:** 2\ **Last updated:** [April 8, 2025, 5:15am UTC](https://community.openai.com/t/realtime-api-error-committing-input-audio-buffer-the-buffer-is-empty/975381 "2025-04-08T05:15:11Z")
I’m trying to basically replicate the functionality of the openai-realtime-console. However, I can’t even get past receiving an audio response back. Currently I’m able to create a session, update it, create a conversati… --- ## [How to forward OpenAI's stream response using FastAPI in python?](https://community.openai.com/t/how-to-forward-openais-stream-response-using-fastapi-in-python/963242)
**Author:** [@aayush\_jerry](https://community.openai.com/u/aayush_jerry)\ **Replies:** 4\ **Last updated:** [March 5, 2025, 8:23pm UTC](https://community.openai.com/t/how-to-forward-openais-stream-response-using-fastapi-in-python/963242 "2025-03-05T20:23:04Z")
This is my code to retrieve stream response from OpenAI’s model which is event based. (I have shown only core part) client = OpenAI(api\_key=OPEN\_AI\_API\_KEY) class EventHandler(AssistantEventHandler): def on\_text\_de… --- ## [Persistent "server\_error" with gpt-4o-realtime-preview Models on response.create](https://community.openai.com/t/persistent-server-error-with-gpt-4o-realtime-preview-models-on-response-create/1135114)
**Author:** [@daniel.donnelly](https://community.openai.com/u/daniel.donnelly)\ **Replies:** 0\ **Last updated:** [March 4, 2025, 11:54am UTC](https://community.openai.com/t/persistent-server-error-with-gpt-4o-realtime-preview-models-on-response-create/1135114 "2025-03-04T11:54:25Z")
Hi OpenAI team and community, I’m using the Realtime API (realtime=v1) and hitting a consistent “server\_error” on every response.create call, whether with text or audio input. This happens with both gpt-4o-realtime-prev… [Next page](https://community.openai.com/tag/python/45.md?match_all_tags=true&page=1&tags%5B%5D=python)