Back to skill

Security audit

Industry Deep Dive Pipeline

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because it requires external searches beyond its declared file permissions, creating a potential disclosure risk; the review is incomplete because some references could not be resolved.

Before installing, consider whether you can authorize external searches and retrieval for this workflow and prevent confidential information from being sent in search queries. The scan examined all 12 listed components, but its overall analysis was partial because some references could not be resolved, so the evidence is incomplete.

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
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

The workflow requires external searches and source retrieval despite declaring only file access permissions.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The workflow mandates external searches and source retrieval while declaring only Read/Write tools and file permissions, leaving required network activity outside its stated authorization boundary. If an agent uses ambient search capabilities to satisfy Step 2, queries derived from confidential source materials could disclose sensitive information; Step 3's explicit disclosure safeguards reduce risk there but do not clearly cover the earlier fact-verification searches. External research is legitimate for this skill, and the text does not demonstrate an actual permission bypass, but the inconsistent scope creates an avoidable disclosure risk.

Content

No source excerpt is available for this finding.