Dev.to Security πŸ” Cybersecurity πŸ‘ 0 πŸ“– 4 min read

Start MeowShield variable renaming at Normal or Hexadecimal, not the most exotic alphabet

Exotic identifier alphabets are an easy place to over-optimize a JavaScript protection build. A file full of punctuation-like names or visually odd glyphs looks more hostile than one whose locals still resemble ordinary

Exotic identifier alphabets are an easy place to over-optimize a JavaScript protection build. A file full of punctuation-like names or visually odd glyphs looks more hostile than one whose locals still resemble ordinary minified identifiers. That visual jump is not a stronger control-flow transform, and it is not a reason to start at the most unusual renaming style MeowShield offers.

I help promote MeowShield. The product's own guidance on Variable Renaming is conservative: start at Normal or Hexadecimal, then move to an exotic style only if you want stronger confusion and have already confirmed that the build still behaves.

Renaming changes names, not bindings

Variable Renaming is a select control, defaulted to Normal. It rewrites identifiers so the resulting file is harder to read, and so familiar naming patterns are less obvious during reverse engineering. That is the documented effect: readability and pattern recognizability.

What it does not do is change the meaning of a binding. A local that used to be called invoiceTotal is still the same local after it is renamed. If you need the execution path itself to be harder to follow, that is a different class of option β€” control-flow protection, not an alphabet change.

It also does not hide secrets. A token, a price you intend to enforce, or an entitlement flag that lives in client JavaScript is still in the client after its surrounding names are rewritten. Anyone who can run the file can still observe the value. Server-side authentication and authorization remain necessary, and no protection is absolute. The goal is to increase analysis cost while preserving the behavior you need.

What the styles actually are

The documented styles are Normal, Hexadecimal, Chinese, Bubble, Punctuation, Special, and Transparent. The Low preset, which is the compatibility-first starting point, uses Normal. That is a deliberate baseline, not an unfinished setting.

If you later call the authorized API from a server or CI job, the same choice is the numeric renameVariables field on the options object. The documented mapping is:

  • 0: Normal (the default)
  • 1: Chinese
  • 2: Bubble
  • 3: Punctuation
  • 4: Hexadecimal
  • 5: Special
  • 6: Transparent

Keep the API token in that server or CI environment, never in public browser code. The full options object is documented at https://meowshield.com/api-docs; do not invent additional fields around this one.

I will not rank the exotic styles by how hard they are to deobfuscate. The public docs do not provide that ranking, and a style that looks more confusing to a human reader is not automatically harder for a tool that already normalizes identifiers. Treat the list as a set of readability strategies, not as a security ladder.

Start at Normal or Hexadecimal

Normal is the honest first measurement. It is what Low already uses, so a first Playground build with the Low preset already includes it. You can compare the protected file against your source, confirm that public exports and contracts still hold, and see what ordinary renaming does to readability without introducing an alphabet your toolchain has never seen.

Hexadecimal is the other recommended starting point. It is still a relatively conventional identifier style in compiled or generated JavaScript, which makes it a smaller compatibility bet than punctuation-heavy or visually unusual names. If you want a clearer break from the original source names than Normal provides, try Hexadecimal next, as a single change, on a file you already know how to test.

Keep public exports and documented contracts stable. A paid SDK or a script that other code loads by name cannot afford a renamed export. Skip vendored libraries unless you have a specific reason to run them through the same renaming pass.

Move to an exotic style only after you can measure it

Chinese, Bubble, Punctuation, Special, and Transparent exist for stronger confusion once you have a baseline that already runs. "Stronger confusion" is a readability claim. It is not a promise that one of these styles is the hardest to reverse, and it is not a substitute for choosing the right file or the right preset.

Change only the renaming style, generate output, and run the same load and behavior tests you already use. Pay attention to anything in your pipeline that assumes identifier shape: a regex in a fragile post-process step, a log sanitizer, a review tool, or a host that inspects function names. If the only result is a break in a tool you still need, go back to Normal or Hexadecimal.

If the build fails more broadly, official troubleshooting still applies: lower the protection level or disable options, then add them back one at a time. If it still fails, contact [email protected] with the configuration, runtime environment, and error details.

A useful sequence is:

  1. Generate a Low baseline, which already uses Normal renaming.
  2. Confirm behavior, compatibility, and that public contracts still resolve.
  3. If you want a more conventional but less readable alphabet, switch only to Hexadecimal and re-test.
  4. Only then consider one exotic style, and keep it only if the extra confusion is worth any friction you measured.

The official path is unchanged: start with a preset in the Playground, understand the option in Docs, then move authorized integrations to the API. Identifier style is a poor place to begin with the most visually aggressive setting.

If you want to compare Normal and Hexadecimal on a real file in the Playground, start at https://meowshield.com.

πŸ“° Read the original article on Dev.to Security

Originally published by Dev.to Security. Aggregated on AIWithGhost for educational purposes β€” full credit and traffic to the original publisher.