ClawHub
Back to skill

Security audit

Text Diff

Security checks across malware telemetry and agentic risk

Overview

This is a simple local text-diff tool with some documentation quality issues but no evidence of hidden, networked, credential-seeking, persistent, or destructive behavior.

Install only if you need a local text-file diff helper. Be aware it does not appear to support Word/PDF parsing or contract-specific semantic analysis despite one scenario claiming that; use plain text inputs or convert documents first.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding
Without declared permissions the skill's intent is opaque and cannot be validated.

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The manifest description at L003 and the main documentation at L034 describe a text-file diff utility using Python's difflib, with unified/context/side-by-side modes. Scenario 4 expands the claimed behavior to Word/PDF/markdown support, colored semantic highlighting, summary counts, and special marking of contract clauses, which is materially broader than the stated skill purpose and implied implementation.

Intent-Code Divergence

Low
Confidence
82% confidence
Finding
L027 says the expected output includes 'Exit code 0' for identical files and 'exit code 1' for differing files, but the file is primarily documented as a Python utility that displays diffs rather than a clearly specified CLI contract. In addition, the Exit Codes section at L104-L106 appears truncated, creating documentation that does not reliably match the stated behavior.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
This markdown file includes multiple user-facing examples and expected outputs in Chinese, but the skill does not indicate that language is configurable or that it is intentionally region-specific. Per the policy, forcing or assuming a specific language without user opt-in is a natural-language policy concern.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.