whats-agent-doing

Privacy Policy

What’s Agent Doing · Effective October 2, 2026

No data leaves your Claude Code

What’s Agent Doing is a Claude Code plugin that shows, in a small box above the prompt, what the agent is doing right now and what it has done. It runs entirely inside your Claude Code session.

• It collects nothing. The plugin has no analytics, telemetry, tracking or crash reporting.
• It sends nothing. The plugin makes no network requests. It has no server, and nothing it sees is sent to its author or to anyone else.
• It stores nothing. The plugin writes no files and keeps nothing between sessions. What it shows lives in the session’s memory and is gone when the session ends.
• It shares nothing. There is no data to sell, rent or hand to third parties.

What the plugin sees, and what it does with it

To say what the agent is doing, the plugin reads the session’s activity as it happens: your prompts, the agent’s tool calls (file paths, URLs, command descriptions, MCP tool names), and, where the model streams it, a line of its thinking. It uses these only to draw the box on your screen. Anyone who can see your screen, or a screen share, can read what the box shows.

The plugin only displays. It never changes, allows or denies anything the agent does, including permission prompts.

Claude Code itself

This policy covers only this plugin. Claude Code, and the model it talks to, handle your data under Anthropic’s own terms and privacy policy, whether or not the plugin is installed. The plugin adds nothing to what Claude Code sends.

Changes

If this policy changes, the new version will be posted on this page with a new effective date. The history of every change is public in the repository.

Contact

Questions about this policy: GitHub Issues.

This site is open source. Improve this page.