Five capture channels feed one AI pipeline. Support gets escalation-ready context. Engineering gets root cause, a code fix, and a regression test. Here’s exactly how it works.
Five ways in, from passive error monitoring to customer self-service
recording. Every channel feeds the same AI pipeline.
A code snippet that passively monitors for errors before customers report them. Console errors, network failures, unhandled exceptions, captured automatically in real-time.
How it works: Add one <script> tag. Sniffer captures JS errors, console errors, and network failures. No customer action required errors appear in your dashboard before anyone reports them.
A floating button inside your app. Customers click it, capture a screenshot or video, describe the issue. Console logs, network errors, and device info are auto-attached.
Agent sends a simple link. Customer records their screen or takes a screenshot directly in the browser no install, no extension, no Sniffer code on their site needed.
How it works: Agent copies link from helpdesk. Customer records via browser-native API. Single-use, auto-expiring. Optional mic narration for context.
Agent sends a link to the customer. When clicked, it unhides the Sniffer capture button on the site because the Sniffer JS snippet is already running, this captures the full technical context.
How it works: Requires Sniffer snippet already installed on the customer’s site. Link activates the capture button for that session. Console logs, network traffic, user actions all captured alongside the recording.
Full capture toolbox for internal teams: screenshots, screen recordings, annotations, user actions, blur tool, pause, countdown timer. Console logs and network requests auto-attached. Works on any website.
Who uses this: Engineering teams, QA testers, and agency PMs reviewing client sites. Install the extension, browse any site, capture and annotate, assign to the right team.
What gets captured (varies by channel)
Extension + JS Snippets + Sniffer Record Link also capture
Every capture feeds into
Regardless of which channel captured the bug, Sniffer runs the same
intelligence pipeline on every issue automatically.
AI watches the recording and auto-generates numbered reproduction steps from video + console + network activity.
Step 1: Navigate to /checkout
Step 2: Click “Apply coupon”
Step 3: ⚠️ TypeError at line 142
AI examines console errors, network failures, and stack traces captured alongside the recording.
TypeError: Cannot read property
‘discount’ of null
→ checkout.js:142
AI identifies the root cause and suggests a code fix with confidence level. Not generic, built from the actual error data.
Root cause: user.discount is null
Fix: Add null check before
payment init · 92% confidence
AI creates a Playwright/Cypress regression test from the recording data, real steps, real inputs, real assertions.
test(‘coupon null check’, …)
→ Push to CI so this bug
never ships again
Every capture feeds into
Support and engineering see different views of the same AI-analyzed bug.
No back-and-forth needed.
Agent’s first interaction is reading the finished AI report, not asking the customer to describe the problem again.
AI Repro Steps : numbered steps ready to paste into Jira
Root Cause Summary : plain-language explanation for the escalation
Error Grouping : "12 other customers hit this same issue"
One-Click Escalation : push to Jira, Asana, or Azure DevOps from your helpdesk
Engineers see the solution and test script, exposed directly to AI coding tools via MCP. No context-switching.
Suggested Code Fix : exact file, line number, confidence score
Regression Test Script : Playwright/Cypress, ready for CI
MCP Context Feed : say "fix this" in Cursor and the AI pulls everything
Full Session Replay : video + console + network + user actions
Sniffer isn’t just a capture tool, it’s your bug lifecycle hub. Built-in task
management, error intelligence, and integrations with the tools you
already use.
Sniffer is the task board, not just a capture tool. Kanban, List, and Table views with severity, assignees, tags, and estimation. No need for a separate project tracker.
“12 other customers reported this same issue”: pattern detection across recordings. Trending Issues dashboard surfaces the bugs hurting the most customers.
Recording links sent from your helpdesk. AI reports flow back into your tickets. One-click escalation to your engineering tools.
Exposes full bug context from all capture channels directly to AI coding tools. Engineers say “fix this bug” and the AI pulls the complete report.
Full UI translation across multiple languages including Arabic for MENA markets. Auto-translate customer descriptions so agents read every report in their own language.
Customer blur tool for PII redaction, auto-detection for credit card numbers and SSNs, single-use recording links, configurable retention, and visible auto-delete countdown.
Five capture channels. One AI pipeline. Full bug lifecycle
coverage from the customer’s screen to the engineer’s code
fix.
