openclaw - 💡(How to fix) Fix [Bug] thought_signature 400 regression in 2026.4.25 – fix was supposedly in 2026.4.24 [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
openclaw/openclaw#72879Fetched 2026-04-28 06:31:01
View on GitHub
Comments
1
Participants
2
Timeline
2
Reactions
1
Author
Timeline (top)
commented ×1subscribed ×1

The thought_signature Google Generative AI API 400 error has regressed in 2026.4.25. This was previously addressed and that issue was closed after the 2026.4.24 release, but the error is back.

Error Message

Google Generative AI API error (400): Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call `def…

Root Cause

The thought_signature Google Generative AI API 400 error has regressed in 2026.4.25. This was previously addressed and that issue was closed after the 2026.4.24 release, but the error is back.

Code Example

Google Generative AI API error (400): Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call `def…
RAW_BUFFERClick to expand / collapse

Summary

The thought_signature Google Generative AI API 400 error has regressed in 2026.4.25. This was previously addressed and that issue was closed after the 2026.4.24 release, but the error is back.

Error

Google Generative AI API error (400): Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call `def…

Context

  • Version: 2026.4.25
  • Model: Google Generative AI (Gemini) — model auto-selected via agent config
  • Trigger: Occurs during tool-calling turns with Gemini models
  • Environment: OpenClaw agent runtime

The fix that shipped in 2026.4.24 is no longer sufficient — the thought_signature is still being dropped or stripped somewhere in the request pipeline for the native Google Generative AI provider.

Related Issues

  • Previous fix in 2026.4.24 (issue closed after that release)
  • #53658 (Google OpenAI-compat fallback + missing thought_signature during replay)
  • #34008 (Gemini 3 missing thought_signature via Ollama Cloud)
  • #33272 (reasoning_effort sending thinking metadata to unsupported endpoints)
  • #509 (thought_signature errors during Antigravity OAuth model fallbacks)

Steps to Reproduce

  1. Configure an agent to use a Google Generative AI (Gemini) model that supports tool calling
  2. Have the model participate in a conversation requiring tool calls
  3. Observe the 400 error: "Function call is missing a thought_signature in functionCall parts"

extent analysis

TL;DR

The thought_signature is being dropped or stripped in the request pipeline for the Google Generative AI provider, causing a 400 error, and a revised fix is needed to ensure its inclusion.

Guidance

  • Review the changes made in the 2026.4.24 release to understand what previously resolved the issue and how it may have regressed.
  • Investigate the request pipeline for the native Google Generative AI provider to identify where the thought_signature is being dropped or stripped.
  • Compare the implementation with related issues (#53658, #34008, #33272, #509) to see if a similar fix can be applied.
  • Consider adding logging or debugging statements to track the thought_signature through the request pipeline to pinpoint the issue.

Example

No code snippet is provided due to the lack of specific implementation details in the issue.

Notes

The issue seems to be specific to the Google Generative AI (Gemini) model and the OpenClaw agent runtime, so any fix may need to be tailored to this environment.

Recommendation

Apply a workaround to ensure the thought_signature is included in the request pipeline, as the previous fix in 2026.4.24 is no longer sufficient. This is recommended because the error is causing degraded model performance and the previous fix has regressed.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

openclaw - 💡(How to fix) Fix [Bug] thought_signature 400 regression in 2026.4.25 – fix was supposedly in 2026.4.24 [1 comments, 2 participants]