Getting request violated model provider acceptance policy

Support Bug Reports
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

I am getting the error “Request violated the model provider’s acceptable use policy Request ID: c86b7cd6-7bfd-4502-a366-9e5876455248”, on the below prompt

Create a file at app/resources/stopwords.txt. Each line should contain one
common English word that spaCy’s built-in English stopword list includes
(the, a, an, is, are, and so on — use spaCy’s actual default list). After
that list, add these words on their own lines: meeting, update, team,
project, report, review, agenda, discussion, notes, summary, call, sync,
deadline, task, status.

Steps to Reproduce

Create a file at app/resources/stopwords.txt. Each line should contain one
common English word that spaCy’s built-in English stopword list includes
(the, a, an, is, are, and so on — use spaCy’s actual default list). After
that list, add these words on their own lines: meeting, update, team,
project, report, review, agenda, discussion, notes, summary, call, sync,
deadline, task, status.

Expected Behavior

…

Operating System

Windows 10/11

Version Information

Version: 3.10.17 (user setup)
VS Code Extension API: 1.125.0
Commit: c89f45b831621cdc5e951694db44fecd8fab1150
Date: 2026-07-05T06:39:45.228Z
Layout: Agent Window
Build Type: Stable
Release Track: Default
Electron: 40.10.3
Chromium: 144.0.7559.236
Node.js: 24.15.0
V8: 14.4.258.32-electron.0
xterm.js: 6.1.0-beta.256
OS: Windows_NT x64 10.0.26200

For AI issues: which model did you use?

sonnet 5 and opus 4.8

For AI issues: add Request ID with privacy disabled

Request ID: c86b7cd6-7bfd-4502-a366-9e5876455248
{“error”:“ERROR_OPENAI”,“details”:{“title”:“Unable to reach the model provider”,“detail”:“Request violated the model provider’s acceptable use policy”,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:true}
Request violated the model provider’s acceptable use policy
[invalid_argument] Error
RetriableError: [invalid_argument] Error
at FBv (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:9066:9691)
at XEr (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:9066:8426)
at ssm (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:9067:11963)
at Object.classify (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:9067:16470)
at $Bv (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:9067:5856)
at async _Ma.runWithSharedTurnRunner (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:9067:15822)
at async wq.runAgentLoop (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:20486:3067)
at async ill.streamFromAgentBackend (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:20514:14006)
at async ill.getAgentStreamResponse (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:20514:23428)
at async uBe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/c:/Users/MuhammadBilalBashir/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.glass.main.js:9231:2399)

Additional Information

…

Does this stop you from using Cursor

No - Cursor works, but with this issue

5

Hey, thanks for the detailed report and the Request ID.

That block is coming from Anthropic’s content moderation filter (the model provider), not from Cursor. We’re just showing their response. Your prompt is totally harmless, it’s a false positive from their classifier, and it can happen sometimes on safe tasks like generating a word list. Your account and billing are fine, and this failed request does not count toward usage.

What you can do:

  • Retry, it’s often transient and works on the second try.
  • Rephrase the prompt or split it into smaller parts.
  • Switch to a non-Claude model (for example, GPT) for this specific request.

Let me know if it keeps happening consistently even after retries and on other models, then we can dig deeper.