Databricks AI Gateway BYOK — Agent mode tool calling errors

Support Bug Reports
, ,
1

Issue 1 — Claude: invalid tool spec (Agent mode, even on simple prompt)

  • Model: system.ai.claude-haiku-4-5
  • Mode: Agent
  • Prompt: say OK (no tools requested by user)
  • Error: INVALID_PARAMETER_VALUE: Missing 'type' in the tool specification
  • Request ID: 91d48792-94ab-46e8-b731-d7196abde708

Looks like Cursor’s tool payload format is rejected by the upstream provider (via Databricks gateway).

Issue 2 — Gemini: missing thought_signature after tool call (Agent mode)

  • Model: system.ai.gemini-3-5-flash
  • Mode: Agent
  • Behavior: Shell tool runs and returns output, then follow-up request fails
  • Error: Function call is missing a thought_signature in functionCall parts (HTTP 400 from provider)
  • Request IDs:
    • 245fbd0e-c4b2-42f4-8a35-cbd6b2ec9714
    • 8840843d-28a1-4c0d-a8e2-eebe15eba6f6

Issue 3 (secondary) — legacy model names via same endpoint

When using databricks-* model names (instead of system.ai.*), provider returns:

  • NOT_IMPLEMENTED: 'databricks-claude-opus-4-6' is no longer available. Use Unity Catalog model services (v3).
  • Request IDs: ed6af070-499d-4d63-adb3-7a56da8226a0, 182b2daf-6bf7-40ef-8e16-ca877dda3e38

The same model 'databricks-claude-opus-4-6' used with different endpoint (workspacedbc-fddc0605-e757) works fine. No errors all works well with models `databrics*`

  • Request IDs: 15e34231-e652-47fc-8200-5a037fff03a2, ab1da977-838d-4d91-9cf0-97c893f8162c, 06d5773e-8f65-4068-aa84-cb265840f2e3
2

Hi there!

We detected that this may be a bug report, so we’ve moved your post to the Bug Reports category.

To help us investigate and fix this faster, could you edit your original post to include the details from the template below?

Bug Report Template - Click to expand

Where does the bug appear (feature/product)?

  • Editor, Tab & Chat (autocomplete, Composer, in-editor agent)
  • Terminal & commands
  • Models, pricing & API keys (availability, Auto/Max, BYOK/Bedrock)
  • MCP & tools
  • Cloud Agents & Automations (cursor.com/agents, scheduled/event)
  • BugBot & Code Review
  • Cursor CLI
  • Cursor Mobile
  • Remote (SSH / Dev Containers / WSL)
  • Account, billing & login
  • Something else…

Describe the Bug
A clear and concise description of what the bug is.

Steps to Reproduce
How can you reproduce this bug? We have a much better chance at fixing issues if we can reproduce them!

  1. …
  2. …
  3. …

Expected Behavior
What is meant to happen here that isn’t working correctly?

Screenshots / Screen Recordings
If applicable, attach images or videos (.jpg, .png, .gif, .mp4, .mov)

Operating System

  • Windows 10/11
  • MacOS
  • Linux

Version Information

  • For Cursor IDE: Menu → About Cursor → Copy
  • For Cursor CLI: Run agent about in your terminal
IDE:
Version: 2.xx.x
VSCode Version: 1.105.1
Commit: ......

CLI:
CLI Version 2026.01.17-d239e66

For AI issues: which model did you use?
Model name (e.g., Sonnet 4, Tab…)

For AI issues: add Request ID with privacy disabled
Request ID: f9a7046a-279b-47e5-ab48-6e8dc12daba1
For Background Agent issues, also post the ID: bc-…

Additional Information
Add any other context about the problem here.

Does this stop you from using Cursor?

  • Yes - Cursor is unusable
  • Sometimes - I can sometimes use Cursor
  • No - Cursor works, but with this issue

The more details you provide, the easier it is for us to reproduce and fix the issue. Thanks!

5

cursor-databricks-issues.txt (16.1 KB)

9

Thanks for the report @Tomasz_Kisielewski! Cursor doesn’t fully support Databricks AI Gateway today, and I’m not aware of any short-term work to change that. I’ve logged the Claude tool-spec and Gemini thought_signature failures for the team!

Issue 3 is on the Databricks side. That workspace (dbc-6375e9c0-d24b) is rejecting the legacy pay-per-token databricks-* names (NOT_IMPLEMENTED / use Unity Catalog model services v3). Your other workspace still serves them. That’s something to take up with Databricks.

11

Agree - it’s not ready yet.

Thanks,

15

Hey @Tomasz_Kisielewski,

Quick update on the Gemini side of this. The missing thought_signature failure after a tool call, when running Gemini 3.x through your Databricks AI Gateway, has been addressed on our side and will be rolled out soon if it hasn’t been already.