Enforce support of enforce_user_param to openai post endpoints

[Sameerlite]

Title

Enforce support of enforce_user_param to openai post endpoints

Relevant issues

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/litellm/ directory, Adding at least 1 test is a hard requirement - see details
• I have added a screenshot of my new test passing locally
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem

Type

🐛 Bug Fix
🧹 Refactoring

Changes

The enforce_user_param setting was being enforced on ALL OpenAI endpoint requests, including:

• GET requests like /v1/models (which don't have request bodies)
• MCP-related calls (which were causing 500 errors)

This caused issues because:

1. GET requests don't have request bodies, so adding a user parameter doesn't make sense
2. MCP calls shouldn't be required to have a user parameter

Modified the enforce_user_param logic in /litellm/proxy/auth/auth_checks.py to:

1. Check HTTP Method: Only enforce the user parameter on POST requests
2. Exclude MCP Routes: Check if the route is an MCP route and skip enforcement
3. Focus on Completion Calls: This ensures user parameter is enforced only on actual LLM completion calls (chat/completions, embeddings, etc.)

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
litellm	Ready	Preview	Comment	Dec 3, 2025 6:50am