Pincushion for Claude Code.

Visual feedback that lands in your IDE. Stakeholders drop pins on your live app; your Claude Code agent reads them via MCP — selector, DOM snippet, screenshot, viewport, thread, project context, and acceptance criteria — and ships the change from inside your editor.

The flow

1. Stakeholder drops a pin on any element of your live app via the Chrome extension
2. You open Claude Code in your repo, run /pins or /implement
3. Claude reads the pin's full work packet via MCP — selector, DOM snippet, screenshot, viewport, thread, project context, and likely files
4. Claude makes the change, calls fix_and_resolve with branch + PR URL
5. Deploy hook auto-resolves the pin on production. Pincushion AI verifies the fix on the next deploy and records the verdict

Connect Claude Code

Start with npx pincushion-mcp setup to sign in, register project URLs, and explicitly choose Claude Code. From inside your project, you can also run:

claude mcp add pincushion -- npx pincushion-mcp --project-dir .

Restart Claude Code, then ask it to call get_actionable_pins. Configuration does not prove the agent is connected: review an existing AI or human finding, approve that exact pin, then ask it to implement it. The Chrome extension is optional and only needed to place new browser pins.

For a Crit report, use npx pincushion-mcp claim <report-token> to resume configuration after authenticated project access is confirmed.

Slash commands available after install

• /pins — show all open pins on this project
• /implement — run the full approved-pin loop (claim → fix → resolve)
• /resolve <id> — claim and close one pin
• /critique <url> — Pincushion AI drops ≤3 UI/copy/a11y pins on the page
• /feedback-summary — project overview

Existing feedback arrives as an agent-ready task with the selector, DOM, screenshot, viewport, and thread in one MCP call.