ClawShield Lite – AI Skill Security Scanner

Security checks across malware telemetry and agentic risk

Overview

This appears to be a benign local pattern scanner, but its “SAFE” output should not be treated as a complete security approval.

This skill looks safe to run as a local, lightweight scanner. Treat its output as advisory: a “SAFE” result only means none of its listed string patterns were found, not that the scanned skill is definitely secure.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI09: Human-Agent Trust Exploitation
Low
What this means

Users may place too much trust in a clean result and miss risks that are not covered by the small pattern list.

Why it was flagged

The scanner uses direct substring matches and labels no-match results as “SAFE,” which is purpose-aligned but could be over-interpreted as a comprehensive security verdict.

Skill content
if pattern in code: ... if total == 0:
        risk_level = "SAFE"
Recommendation

Use this as a quick first-pass checker only, and review any skill manually or with a more complete security process before installing.