Free HAR Analytics Practice Kit (Perunlight)

What it does

What this kit is and who it’s for The Free HAR Analytics Practice Kit from Perunlight is a downloadable bundle for QA and analytics engineers who want a local workflow to learn and test HAR-based checks for GA4-style analytics events. It’s aimed at teams that need a safe, offline way to reproduce comparisons without uploading real network captures. Included practice kit and fixtures The download includes an MIT-licensed Node CLI (har-diff.mjs), a worked lesson, test files and four fictional .har captures so you can run example comparisons out of the box. The fixtures let you verify the tool’s output without using sensitive or production HARs. Local CLI: parsing and example-driven learning The Node CLI runs locally with Node.js and focuses on GA4-style single-event-per-request payloads. The README documents tested Node versions and file limits; the kit’s example checks show how the CLI parses events, reports differences, and keeps URLs and parameter values out of reports. MuninCompare: browser GUI for interactive review Perunlight documents a paid, GUI alternative named MuninCompare that reads two HAR files in the browser, recognises GA4 and Adobe Analytics requests, groups events and lists removed, added, changed or repeated sends. The guide includes a demo video and an illustrated manual; installation is via an unpacked Chrome extension. Redaction, export and privacy notes Both the CLI and the browser parsing emphasize local operation and redaction: parsers can redact headers, URLs and common identifiers and export a redacted HTML report. The guide warns redaction is not full anonymisation and recommends reviewing reports before sharing. Scope, limits and what this kit does not do The practice kit has a narrow scope: the free CLI focuses on GA4-style requests and enforces practical file limits (example: 8 MiB / 20,000 entries in the kit). The tools analyse HAR network evidence only and do not validate consent handling, backend delivery, platform-side conversion accuracy or full anonymisation of captured data.