Screenshot to Webhook - Annotate & Send

Screenshot, annotate, send to a webhook. Bugs, client change requests and vibe-coding tasks into n8n, Jira or your AI coding agent.

Screenshot to Webhook turns what you see in the browser into a structured request for humans and AI coding tools. Capture the tab, mark it up with arrows, numbered pins, text and redaction, then deliver the annotated image plus page metadata (page address, title, viewport, browser) to any webhook endpoint you choose: n8n, Zapier, or your own automation. No account, no backend, nothing leaves your machine until you press Send. Built for three jobs: 1. Manual bug detection. AI writes most of the code now; humans still find the bugs by looking. Spot it, grab it, flag it, and the ticket — with the image and the exact page URL — is created by your automation or picked up by the tool that fixes it. 2. Customer change requests. Clients and stakeholders mark up the live site and say what they want changed. The request lands in your backlog as a precise visual ticket instead of a vague email. 3. Vibe coding. Building with Claude Code, Cursor, or Codex? Show it what the app should look like, so the next run starts from a picture and a note instead of a paragraph of prose. CAPTURE Click the toolbar icon and the tab you are looking at is captured instantly, pixel for pixel. No picker, no recording, nothing runs on the page. ANNOTATE Mark up the capture in a full-tab editor: - Arrows, rectangles, and circles - Numbered pins, each with its own note - Text labels - Freehand pen - Pixelate to redact passwords, emails, or anything private before it leaves your machine - Free-text note attached to the entire delivery DELIVER Paste a URL in the settings page and the extension posts the annotated PNG plus structured metadata (page address and title, per-pin notes, time, browser, OS, screen and viewport size) straight to it. Choose JSON with a base64 image or multipart/form-data with a real PNG file part. Keep several destinations side by side, each with its own name, URL, payload format and authentication (Bearer token, API key header or Basic auth, masked in the settings page) plus any custom headers. The editor shows where the report goes and lets you switch destinations before sending; the last one used stays selected. Ready-made recipes in the settings page show how to receive the payload in Zapier, Make, and n8n, and how to decode the image, so reports can be routed into whatever tool your automation platform connects to. MADE FOR AI-ASSISTED DEVELOPMENT Every delivery carries the PNG, the page address and title, each pin's text, and the free-form description as structured JSON (or multipart with a real file part), so your automation — or the LLM behind it — has everything a good ticket needs without parsing an email. Create an issue in your tracker, notify your team, or feed the queue your coding agent works through. Setup takes one paste: the settings page has a "Copy prompt for your coding agent" button that hands over the payload spec and asks the agent to build the receiving webhook and reply with its URL and auth header. PRIVATE BY DESIGN - No account, no sign-in, no servers of our own - Your reports go only to the destination you choose, and only when you press Send - Settings, including tokens, are stored on your device (chrome.storage.local), nowhere else - No analytics, no tracking, no ads WHO IT IS FOR Developers and QA reviewing AI-generated apps, agencies and product teams collecting customer change requests, vibe coders working with Claude Code or similar tools, and anyone who wants what they see in the browser to land in their own systems automatically. PERMISSIONS - "storage": save your webhook destinations - Host access: used only to post your report to the endpoint you choose. The extension never reads or modifies web pages.