# Send feedback from the Vercel MCP

Source: [Vercel](https://vercel.com/changelog/send-feedback-from-vercel-mcp)  
Feed7 permalink: https://feed7.dev/p/send-feedback-from-vercel-mcp-0xey0ti  
Published: 2026-10-23T20:53:00.000Z  
Trust: Official Source (official_source)

## Why Included

Vercel MCP adds submit_feedback so agents can send structured bug, friction, and feature reports. It requires sign-in and only transmits feedback after an explicit user request.

## Source Summary

Vercel MCP now includes **submit_feedback** for bug reports, friction reports, and feature requests. Agents can attach the task, expected behavior, actual result, and relevant tool context.

## Practical Implication

Builders can use it to report failures without leaving an agent session, but should inspect what context is attached and make the request explicit.

## Agent-Ready Context

Vercel MCP now includes **submit_feedback** for bug reports, friction reports, and feature requests. Agents can attach the task, expected behavior, actual result, and relevant tool context.

Builders can use it to report failures without leaving an agent session, but should inspect what context is attached and make the request explicit.

The tool **requires sign-in** and sends feedback only when the user asks. The material does not describe retention, redaction, or response handling.

## Connected Context

Feed7 judgment across 898 accumulated Signals:

This extends Vercel’s structured in-session feedback path from terminal workflows into MCP clients. The important boundary is consent: authentication enables the tool, but an agent should submit only on an explicit user request and after making the attached task and tool context visible. Unspecified retention and redaction leave context handling as an implementation concern.

- [Send feedback from the Vercel CLI](https://feed7.dev/p/send-feedback-from-vercel-cli-0mhjuec) — This is the closest counterpart: both capture task, expectation, and failure context, while the new release makes that reporting pattern available through MCP rather than only the CLI.
- [Vercel Connect now supports TanStack AI](https://feed7.dev/p/vercel-connect-tanstack-ai-1owhrob) — The TanStack integration reinforces the authorization boundary around remote MCP actions: sign-in, per-user credentials, and consent should be resolved before a model invokes an external reporting tool.

## Context Map

- Layer: tools
- Domains: coding
- Topics: mcp, tool-use, dev-ux

## Uncertainty

- The tool **requires sign-in** and sends feedback only when the user asks. The material does not describe retention, redaction, or response handling.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.