このエラーは、Claude Code 側のトランスクリプト再構築で advisor_tool_result と server_tool_use の対応が崩れたときに出ているようです。少なくとも issue 報告では、advisor_tool_result の前に対応する server_tool_use が必要なのに、それが見つからない／ずれる、という形で説明されています。 同じ症状として報告されて

まず自分のケースかどうかを切り分けるなら、親エージェントや対話画面に出る文言だけで判断しないほうがいいです。報告されている症状では、実際の upstream は認証失敗なのに、親側には Request interrupted by user for tool use とだけ見えることがあります。これは subagent を Agent tool で起動しているときに起きた例として 70958 に残

このエラーは、Claude Code 側のトランスクリプト再構築で advisor_tool_result と server_tool_use の対応が崩れたときに出ているようです。少なくとも issue 報告では、advisor_tool_result の前に対応する server_tool_use が必要なのに、それが見つからない／ずれる、という形で説明されています。 同じ症状として報告されて

Getting the “API Error 529 (Overloaded)” in Claude? Don’t panic. Here’s a complete step-by-step guide to understand the issue, apply quick workarounds, and get back to work fast. If you are currently using Claude Code to ship a project or calling the Anthropic API for a critical task and suddenly see a string of retries in your terminal, don’t panic…

Official Flutter/Dart SDK for SabPaisa Payment Gateway 2.0 — payments, refunds, transaction enquiry, webhook verification. Changelog excerpt: - Initial release - Payment session creation with checksum - Return URL verification - Transaction enquiry - Refund creation, status, and listing - Webhook signature verification with replay protection - Full error hierarchy (SabPaisaError, ApiError, ChecksumError, ValidationError) - Retry with exponential backoff for safe/idempotent requests - Request timeout (30s default) - Timing-safe HMAC comparison

Error is still there occasionaly. So we will also try to go back to the copmpletions API, although it is clear by now that is something on Azure’s side. Something with the images seems to be very unstable.

as expected the problem is not there anymore, which confirms that the issue was either at azures side in terms of server hosting or on openais side with regard to the model version they deployed there. I guess we will not find out as such kind of bugs will never be openly reported.

We are experiencing the exact same issue. Switching to a different region or using the completions API instead of the responses API seems to fix it. Do you have any new information on it?

hi thanks for support. I was posting the problem I have here, because the error is raised in the openai package. but thanks for pointing out about the mime types as well as the strategy on how to tackle this issue even more systematically.

OpenAI is not where you’d report problems with Azure AI services. You get to pay **Microsoft** to tell them they are failing you. learn.microsoft.com ### Foundry Tools support and help options - Foundry Tools Learn how to obtain help and support for questions and problems when you create applications that integrate with Foundry Tools. You might try to re-deploy, perhaps to a different region. Then explore making the similar vision request by direct API call outside of LangChain or SDK to see if it is the particular request being sent. Ensure you are using a MIME type in the base64 data URL that is the same as the actual image file (while OpenAI will even let you just send “`image/`” with no type.)

Hi, I’m facing the same issue: 2 of 3 Dalle-3 requests have this error, no logs available to check what’s wrong and the support did not help. Please, let me know if you find the root cause.

Same issue - getting 424 when using a successfully deployed MCP server (streamable HTTP, app deployed on Databricks) Responses from OpenAI Agents SDK / responses API: * APIError: Error retrieving tool list from MCP server: ‘exa-mcp’. Http status code: 424 (Failed Dependency) * ERROR:openai.agents:Error getting response: Error code: 424 - {‘error’: {‘message’: “Error retrieving tool list from MCP server: ‘exa-mcp’. Http status code: 424 (Failed Dependency)”, ‘type’: ‘external_connector_error’, ‘param’: ‘tools’, ‘code’: ‘http_error’}}. Using the same MCP tool config through Anthropic works fine