agentsclimarketplace
Sign in

Extracting and classifying strings

Skill meltedinhex/analyst-ai-pack/skills/extracting-and-classifying-strings

Extracts ASCII and Unicode strings from a binary and classifies them into investigative categories: URLs, IPs, file paths, registry keys, mutexes, commands, and API names, prioritizing the analyst's attention. Activates for requests to extract strings, classify strings output, or pull human-readable artifacts from a sample.From its SKILL.md

Install
npx -y skills add meltedinhex/analyst-ai-pack --skill extracting-and-classifying-strings

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

2.9 KB, 488 tokens by cl100k_base, as published. Nobody here has run it

Extracting and Classifying Strings

When to Use

  • You are triaging a sample and want quick investigative leads from its readable strings.
  • You need both ASCII and UTF-16LE (Unicode) strings, which many tools miss by default.
  • You want strings grouped by meaning (URLs, paths, registry, commands) rather than a flat dump.

Do not use plaintext strings as the whole story for packed/obfuscated samples — the useful strings may be encrypted; pair with entropy checks and deobfuscation (FLOSS) when the dump is sparse.

Prerequisites

  • Python (stdlib) or strings/FLOSS; the sample handled inertly in the lab.

Safety & Handling

  • Extract strings from the inert file; never execute the sample.
  • Defang any URLs/IPs before putting them in a report or ticket.

Workflow

Step 1: Extract ASCII and Unicode strings

Pull printable runs at a minimum length (default 4–5) for both ASCII and UTF-16LE encodings to avoid missing Windows wide strings.

python scripts/analyst.py strings sample.bin --min-len 5

Step 2: Classify by category

Bucket strings into URLs, IPs, domains, file paths, registry keys, mutex/event names, suspicious commands, and API names so leads surface immediately.

Step 3: Prioritize leads

Promote network indicators, command lines, and suspicious API references to the top; defang network artifacts for safe handling.

Step 4: Recognize obfuscation

A near-empty or junk-only dump signals packing/string obfuscation; route to entropy analysis and FLOSS/manual deobfuscation.

Validation

  • Both ASCII and Unicode strings are extracted (Windows wide strings are not missed).
  • Strings are grouped into useful categories, not a flat list.
  • A sparse/garbage dump is correctly read as an obfuscation signal, not "clean".

Pitfalls

  • Extracting ASCII only and missing UTF-16LE strings.
  • Treating a packed sample's empty dump as benign.
  • Reporting live URLs/IPs without defanging.

References

What ships with it: 3 files

4.1 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.