Stack trace classification — Generic / Headless / Custom
Stack trace classification — Generic / Headless / Custom
Use this file when the platform isn't Shopify, Magento, WooCommerce, or BigCommerce — or when the store runs a headless or custom frontend where platform-specific path patterns don't apply.
Classification approach
Without reliable path patterns, classify by what you can identify:
Known telemetry scripts (discard)
Non-actionable — discard silently:
connect.facebook.net/— Meta Pixelwww.google-analytics.com/— Google Analyticsstatic.ads-twitter.com/— Twitter/X Pixelanalytics.tiktok.com/— TikTok Pixelsnap.licdn.com/— LinkedIn Insight Tagsc-static.net/— Snapchat Pixelbat.bing.com/— Microsoft/Bing Adscdn.heapanalytics.com/,heapanalytics.com/,heap-api.com/— Heap (CDN + API endpoints)js.posthog.com/— PostHogcdn.segment.com/— Segmentcdn.amplitude.com/— Amplitude
Known vendor scripts (share with vendor)
Regardless of platform, these are always third-party and not fixable by the merchant:
static.klaviyo.com/— Klaviyogoogletagmanager.com/— Google Tag Managercdn.shopify.com/shopifycloud/— Shopify platform (even in headless)- Any recognisable marketing, reviews, support, or payments CDN
ERR_TYPE as primary classifier
When stack trace origin is unclear:
- JS error types (TypeError, ReferenceError, RangeError, etc.) — likely code-level, treat as fixable candidate
- HTTP error types (any 4XX or 5XX) with no JS stack trace — infrastructure or platform, classify as Platform and discard
Frames on the store's own domain
If a frame points to the store's own domain or a CDN that clearly serves the merchant's app (e.g. Vercel deployment URLs for a known merchant project), treat as merchant code and fixable.
When uncertain
If you can't confidently classify an error, surface it as a fixable candidate but note in the card description that the origin is unclear and the fix may require developer investigation.
Editing this file
Add platform-specific patterns here as you identify them. This file is meant to be extended.