agentsclimarketplace

Triage suspicious content

Skill gaelic-ghost/socket/plugins/cybersecurity-skills/skills/triage-suspicious-content

The Source for macOS Agent Workflows

Install
npx -y skills add gaelic-ghost/socket --skill triage-suspicious-content

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

  • 6 stars6 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 author says it does

Copied from the file, not written here

Safely classify suspicious files, archives, installers, packages, scripts, documents, configuration profiles, browser extensions, URLs, QR codes, messages, and nested payloads before execution. Use when someone receives or discovers sketchy content and needs to know what it is, what active behavior it may contain, and the smallest safe next analysis step.

SKILL.md

2.3 KB, 365 tokens by cl100k_base, as published. Nobody here has run it

Triage Suspicious Content

Overview

Identify the content and its active surfaces without opening, installing, importing, previewing, or following it through a handler that may execute code. Preserve the original and route the smallest useful next check.

Read references/content-preflight.md for format-specific active-content clues.

Workflow

  1. Preserve intake context.

    • Record sender/source, delivery channel, claimed purpose, filenames, timestamps, quarantine metadata, URLs as text, and why it looked suspicious.
    • Hash files and work from a copy.

  2. Identify containers before content.

    • Determine file type from bytes and structure, not extension alone.
    • List archive/package members without broad extraction and note traversal paths, symlinks, nested archives, password protection, or misleading double extensions.

  3. Inventory active surfaces.

    • Look for executables, scripts, macros, embedded files, links, forms, JavaScript, launch/install metadata, configuration payloads, extension permissions, shortened/redirecting URLs, and QR destinations.
    • Treat preview generators and importers as parsers that may be vulnerable.

  4. Check local provenance.

    • Record signatures, signer identity, notarization/quarantine evidence, package receipts/metadata, document producer, URL host and certificate context, and expected vendor distribution channel.
    • Do not infer trust from any one field.

  5. Route without execution.

    • Use reputation checking for hash/domain/vendor context, script/document analysis for active text, static malware analysis for capabilities, and reverse engineering for binary internals.
    • Select isolation before any dynamic behavior.

Output

Return intake identity, actual type/container, active surfaces, provenance observations, suspicious and benign explanations, confidence, and the first safe next workflow.

Keep looking

Skills are one crate of 328,083. 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.