BugDump is a bug reporting widget for websites. You add one script tag, a floating bug button appears in the corner of your app, and from that moment on your users can file bug reports that arrive with the context you actually need to fix them: console logs, network requests, a session replay, environment details, and an annotated screenshot.
This post explains what BugDump does, what a report looks like, and where it fits in your workflow.
The problem it solves
Most bug reports from real users look like this: "The page doesn't work." No URL, no browser version, no steps, no error message. You reply asking for details, the user has moved on, and the bug sits unfixed until someone on your own team happens to hit it.
The information you needed existed at the moment the bug happened — in the console, in the network tab, in the DOM. It just wasn't captured. BugDump's job is to capture it automatically, so a one-line complaint becomes a complete case file.
How the setup works
Installation is a single line of HTML:
<script src="https://bugdump.com/sdk/latest.js" async data-api-key="your-api-key"></script>
The script auto-initializes from the data-api-key attribute — no SDK calls, no build step, no framework requirements. It works with React, Vue, or vanilla JavaScript, because it doesn't touch your application code at all. If you prefer a bundled dependency, there is an npm package (@bugdump/sdk) with full TypeScript types, and a manual init mode for programmatic control.
The widget renders inside a Shadow DOM, so its styles never leak into your app and your CSS never breaks the widget. You can set the theme (light, dark, or follow the OS), change the trigger icon, reposition or hide the button entirely, and override every UI string — useful if your product isn't in English.
What's inside a report
When a user clicks the bug button, describes the problem, and hits send, the report that lands in your dashboard includes, automatically:
Console logs — the last 300 entries (2,000 while recording), with errors and warnings highlighted, searchable.
Network requests — the last 150 requests (1,000 while recording) with status codes; failed requests are flagged. Request and response bodies can be included if you opt in.
Session replay — rrweb-based DOM snapshots of what the user did leading up to the bug, scrubbable like a video.
Environment — browser, OS, device, screen size, viewport, and the page URL.
Performance metrics — TTFB, FCP, DOM-ready time, load time, and JS heap usage.
On top of the automatic context, users can attach a screenshot (with annotation tools: arrows, rectangles, freehand drawing, text, and blur for hiding sensitive data), a screen recording, a voice note, or files. None of it is required — the minimum viable report is a sentence and a click, and the telemetry still comes along.
If your app has logged-in users, bugdump('identify', { email }) attaches the reporter's identity so you know who hit the problem without asking them to type their email.
From report to fix
Reports land in a dashboard where they become tasks on a kanban board. From there:
Integrations forward bugs to the tools you already use — Telegram, Slack, Discord, GitHub Issues, Jira, Asana and more on paid plans. On the free plan, new reports reach you by email and in the app.
The MCP server connects your project to Claude Code, Cursor, or any MCP-capable client, so an AI assistant can read a bug's full context — reports, console, network, screenshots — and investigate it in natural language.
A public portal (optional) lets reporters view reports, and you can publish a roadmap and changelog alongside it.
What it costs
Pricing is based on how many reports you collect — not seats, and there is no "contact sales" tier.
Free — $0 forever: 500 reports a month per project, 1 GB storage, 3 projects, email notifications, the Claude Code MCP server, and the full capture stack (screenshot, replay, console, network).
Pro — $4.99/month: 10,000 reports a month per project, 10 GB storage, 20 projects, the full integration list, webhooks, and the option to remove BugDump branding from the widget.
Ultra — $9.99/month: 100,000 reports a month per project, 50 GB storage, 100 projects, and the highest rate limits — built for high-volume products and agencies running many client sites.
Prices exclude tax; Paddle, which handles billing, adds any tax due at checkout. The free plan doesn't require a credit card, so you can evaluate the whole flow — widget, report, dashboard — before paying anything.
Who it's for
BugDump is deliberately small-scale friendly. If you run a side project, an indie SaaS, or a small product team, you probably don't want to configure a heavyweight issue tracker just to hear about broken buttons. One script tag and a kanban board is the whole system.
It also fits agencies and teams running beta programs: non-technical reporters never see a form asking for "browser version" or "steps to reproduce" — they click a button and write a sentence, and the widget gathers the technical half of the report for them.
Where to start
Create a free account, create a project, and paste the script tag into your site. The button appears immediately, and the first report you send yourself is the fastest way to see what your users' reports will look like. If you want to go deeper, the SDK supports feature toggles, noise filters for chatty third-party scripts, custom translations, and programmatic control — but the default configuration is designed to be the one most sites ship.