Codex Desktop 26.810.41047 returns 400: input[0].tools[0].description is empty with Azure Responses API
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.5k
- PR merge metrics
- PR metrics pending
Description
What version of the Codex App are you using (From “About Codex” dialog)?
Desktop version: 26.810.41047 Runtime/client version: 0.148.0
What subscription do you have?
Provider: Azure OpenAI / Foundry
What platform is your computer?
Windows
What issue are you seeing?
Invalid 'input[0].tools[0].description': empty string. Expected a string with minimum length 1, but got an empty string instead.
What steps can reproduce the bug?
After updating to Codex Desktop 26.810.41047, requests sent through a custom Azure OpenAI / Microsoft Foundry provider fail with HTTP 400.
The API rejects a dynamically generated tool because its description field is an empty string
What is the expected behavior?
No response
Additional information
Desktop version: 26.810.41047
Runtime/client version: 0.148.0
Provider: Azure OpenAI / Foundry
Endpoint: /openai/v1/responses
Error: input[0].tools[0].description empty_string
First occurrence: 2026-08-14 04:58:32
Last observed: 2026-08-14 07:57:46
The issue is reproducible and has occurred 18 times in the local logs.
Observed failures by model:
gpt-5.6-sol: 10
gpt-5.6-luna: 7
gpt-5.6-terra: 1
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the failure with Codex Desktop 26.810.41047, Runtime/client 0.148.0, and a custom Azure OpenAI / Microsoft Foundry provider at /openai/v1/responses. Inspect the generated request and local logs around input[0].tools[0].description, using the listed model cases; done means the request is accepted without the empty-description 400 error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- azure, rust
- Domain
- api, backend
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 55/100