ClawHub
Back to skill

Security audit

Resume To Tags

Security checks for vulnerabilities and agentic risk

Overview

This skill is mostly a resume tagging workflow, but it handles sensitive resume data and contact details more broadly than its tag-only description makes clear.

Review this carefully before installing. Use it only when you have permission to process the resumes, understand that resume content may be prepared for LLM processing and Feishu storage, and are comfortable with the helper script extracting phone and email details unless you modify or avoid that script output.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding
The skill claims to produce a pure tag matrix, but the finding indicates it also extracts contact details such as phone numbers and email addresses and does not actually perform several advertised steps. This mismatch can cause users to disclose resumes under false assumptions, leading to unexpected processing of personal data and possible downstream privacy or compliance issues.

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The skill is described as producing a pure tag matrix, but it also extracts phone numbers and email addresses from resumes and includes them in output. This creates unnecessary collection and propagation of personally identifiable information, increasing privacy exposure if outputs are logged, stored, or forwarded to other systems.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
Resumes commonly contain sensitive personal data, and the skill explicitly describes sending that data into external Feishu records without any warning, consent flow, or minimization guidance. In this context, silent transmission of candidate data to a third-party service creates significant privacy, confidentiality, and regulatory risk.

Missing User Warnings

Medium
Confidence
82% confidence
Finding
The script can write structured output derived from resumes to a user-specified file without any privacy controls, redaction, or warning. Because resume text and extracted contact information may be sensitive, saving this data can leak PII through insecure file locations, shared workspaces, backups, or logs.

Static analysis

No suspicious patterns detected.