ClawHub
Back to skill

Security audit

sense-privacy-guard

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed local privacy scanner that may block file operations, with no evidence of exfiltration, persistence, or destructive behavior.

Use this only if you want strict local keyword-based blocking before file operations. Expect false positives, especially because the script also blocks company-confidential terms. Install dependencies in a virtual environment and consider pinning reviewed versions before scanning untrusted PDF, DOCX, or XLSX files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (6)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill instructs the agent to read and inspect arbitrary local files via a scanning script, but it does not declare corresponding permissions. This creates a capability/permission mismatch: users or orchestrators may not realize the skill can access file contents, undermining transparency and policy enforcement even if the stated goal is protective.

Unpinned Dependencies

Low
Category
Supply Chain
Content
pdfplumber
python-docx
openpyxl
Confidence
98% confidence
Finding
The dependency is unpinned, which makes builds non-reproducible and can cause the skill to install different versions over time, including newly introduced vulnerable or malicious releases. In a privacy-focused skill that parses untrusted document formats, dependency drift materially increases supply-chain risk because parser behavior and security posture can change without review.

Unpinned Dependencies

Low
Category
Supply Chain
Content
pdfplumber
python-docx
openpyxl
Confidence
99% confidence
Finding
The package is unpinned and is also separately flagged with known XXE-related advisories, so leaving it unspecified increases the chance of resolving to an unsafe version in some environments. For a skill that inspects DOCX files, this is especially dangerous because DOCX is a ZIP/XML-based format and XML parser flaws can be triggered by attacker-controlled documents.

Unpinned Dependencies

Low
Category
Supply Chain
Content
pdfplumber
python-docx
openpyxl
Confidence
99% confidence
Finding
The dependency is unpinned and also has a known XXE advisory in older versions, so the environment may install a version with XML entity handling issues. Because this skill processes XLSX files, an attacker could supply crafted spreadsheet content that targets XML parsing weaknesses in the library stack.

Known Vulnerable Dependency: python-docx — 2 advisory(ies): CVE-2016-5851 (Improper Restriction of XML External Entity Reference in python-docx); CVE-2016-5851 (python-docx before 0.8.6 allows context-dependent attackers to conduct XML Exter)

High
Category
Supply Chain
Confidence
97% confidence
Finding
python-docx versions before 0.8.6 are affected by an XML External Entity issue that can allow malicious DOCX content to trigger unintended external entity resolution during parsing. In this skill, which explicitly performs deep inspection of potentially sensitive and attacker-supplied documents before read/search/send operations, that increases the risk of file disclosure, SSRF-style access, or parser abuse in a high-value context.

Known Vulnerable Dependency: openpyxl — 2 advisory(ies): CVE-2017-5992 (Improper Restriction of XML External Entity Reference in Openpyxl); CVE-2017-5992 (Openpyxl 2.4.1 resolves external entities by default, which allows remote attack)

High
Category
Supply Chain
Confidence
97% confidence
Finding
Affected openpyxl versions resolve external entities by default, enabling XXE attacks via crafted XLSX content. Since this skill is designed to inspect spreadsheets containing contracts, transaction data, and personal/customer information, the context makes exploitation more dangerous because the parser is exposed to exactly the kinds of untrusted office documents that could be weaponized to exfiltrate local data or reach internal resources.

VirusTotal

58/58 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.