ChatGPT Vercel connector does not expose Vercel MCP runtime logs tool

What issue are you seeing?

Vercel recently announced that agents can access runtime logs through Vercel MCP:

https://vercel.com/changelog/agents-can-now-access-runtime-logs-with-vercels-mcp-server

Vercel’s MCP documentation now lists a get_runtime_logs tool that can retrieve runtime logs for a project or deployment, including filters for project, deployment, environment, level, status code, source, time range, full-text query, and request ID.

However, the ChatGPT Vercel connector/MCP integration currently appears to expose build/deployment logs only, not runtime logs.

What steps can reproduce the bug?

ChatGPT can access Vercel deployment/build logs, but runtime logs are not exposed as an available tool.

What is the expected behavior?

ChatGPT should be able to call Vercel MCP get_runtime_logs for a project/deployment so coding agents can inspect preview/production runtime errors, console output, request failures, and function logs.

Additional information

Why this matters:
Runtime logs are essential for agentic debugging. Build logs help debug failed deployments, but they do not help diagnose production or preview runtime errors after deployment.

Question:
Is ChatGPT using the live Vercel MCP server manifest from https://mcp.vercel.com, or a curated connector/tool wrapper that needs to be updated? Please expose Vercel MCP get_runtime_logs in the ChatGPT Vercel connector.