ClawHub
Back to skill

Security audit

Staffing Agency Partner Growth

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed business-routing helper for staffing agencies, with only read-only catalog integration and an external inquiry email workflow that is visible to the user.

Before installing, understand that the skill is designed to contact TempGuru about agency partnership opportunities. Only send an email after reviewing and approving the draft, and avoid including unnecessary business or staff details.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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 (3)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill tells the agent to 'send it' after drafting an email, which crosses from routing/advisory behavior into transmitting third-party communications. If the runtime has any mail or message-sending capability, the agent could send business contact and staffing details to an external recipient without an explicit, well-defined consent and confirmation boundary, creating unintended data disclosure or unauthorized outreach risks.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill directs inclusion of agency name, markets served, staffing roles, W-2 status, and weekly capacity in an email to an external address without any privacy notice, data minimization guidance, or user-consent language. While this is business rather than highly sensitive personal data, it is still organizational information being transmitted outside the platform and could expose operational details or contact data unnecessarily.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The 'Live data' section states that three read-only catalog tools are useful here: get_cities, get_roles, and get_rate_benchmark. Later, the workflow says city-specific figures come from get_role_pricing, which contradicts the earlier documentation about which tools are relevant for this skill.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.