MD Feedback FAQs

Question 1

What are 'gates' and 'handoffs' in MD Feedback?

Accepted Answer

'Gates' are automated progress indicators that track task completion based on your annotations, showing, for example, '3 fixes remaining' or 'All done'. 'Handoffs' are a feature that preserves the development context for AI agents across sessions, allowing them to pick up seamlessly where they left off.

Question 2

What is MD Feedback?

Accepted Answer

MD Feedback is a VS Code extension and MCP server that empowers developers to provide structured feedback (highlights, fixes, questions) on markdown plans for AI agents to automatically implement, answer, and track, streamlining 'vibe coding' workflows.

Question 3

Can I use MD Feedback if my AI agent doesn't support MCP?

Accepted Answer

Yes, you can still use MD Feedback to annotate your markdown plans within VS Code. For agents not supporting MCP, you can use the 'MD Feedback: Export' command to generate a structured file that your AI agent can then process manually.

Question 4

How does MD Feedback automate AI agent plan implementation?

Accepted Answer

You annotate markdown plans directly within VS Code. An integrated Model Context Protocol (MCP) server allows compatible AI agents (like Claude Code, Cursor, Copilot) to automatically read these annotations, implement the requested changes, and mark tasks complete without any manual export steps.

Question 5

Which AI coding tools are compatible with MD Feedback?

Accepted Answer

MD Feedback supports integration with leading AI coding tools such as Claude Code, Cursor, and Copilot. It's compatible with any AI agent that can leverage the Model Context Protocol (MCP) for direct annotation access.

MD Feedback

MD Feedback

Key Features

Use Cases

Key Features

Use Cases