agentsclimarketplace
Sign in

Deobfuscating malicious javascript

Skill meltedinhex/analyst-ai-pack/skills/deobfuscating-malicious-javascript

Deobfuscates malicious JavaScript from droppers, web pages, and HTA/scriptlets: unrolling string-array and char-code encodings, resolving eval/Function chains, and statically recovering payloads and URLs without executing untrusted code. Activates for requests to deobfuscate JavaScript, decode obfuscated JS, or analyze a malicious script dropper.From its SKILL.md

Install
npx -y skills add meltedinhex/analyst-ai-pack --skill deobfuscating-malicious-javascript

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 21 stars21 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.2 KB, 564 tokens by cl100k_base, as published. Nobody here has run it

Deobfuscating Malicious JavaScript

When to Use

  • You have obfuscated JS (from a phishing page, HTA, .js dropper, or scriptlet) and need its real behavior.
  • You must recover hidden URLs, dropped commands, or a next-stage payload.
  • You want to statically unroll common obfuscation rather than run untrusted code.

Do not use a real browser or node to execute the script for analysis — that runs the malware. Use a sandboxed interpreter (box-js) or static transformation only.

Prerequisites

  • A safe analysis approach: static decoding, or a JS malware sandbox (box-js) in the lab.
  • Familiarity with common obfuscation: string arrays, String.fromCharCode, hex/unicode escapes, eval/Function chains, packers.

Safety & Handling

  • Treat the script as live code; never execute it outside an isolated sandbox.
  • Defang recovered URLs and store dropped payloads password-protected.

Workflow

Step 1: Normalize and de-pack

Pretty-print the source and identify the obfuscation style (array-shuffle, eval-packer, charcode). Decode static encodings first: hex/unicode escapes and base64 literals.

python scripts/analyst.py decode dropper.js

Step 2: Unroll string arrays and char-code builds

Reconstruct strings built from arrays/fromCharCode/concatenation to reveal API names, URLs, and commands.

Step 3: Resolve eval/Function indirection — safely

Replace eval/new Function with logging (or a sandbox) so the constructed code is captured as data, not executed, then recurse on the recovered layer.

Step 4: Extract behavior and IOCs

Identify the dropper actions (WScript.Shell, ActiveX, fetch/XHR), recover URLs and dropped paths, defang, and route payloads onward.

Validation

  • Static encodings (hex/unicode/base64) are fully decoded.
  • String-array/charcode constructions are unrolled to readable strings.
  • eval/Function layers are captured as data and recursed, with no untrusted execution.

Pitfalls

  • Running the script to "see what it does" and infecting the analysis host.
  • Stopping at one layer when the dropper nests several.
  • Missing environment-keyed branches (only acts on certain dates/locales) during static review.

References

What ships with it: 3 files

4.3 KB alongside SKILL.md, 1 of them executable

references/

scripts/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.