ClawHub
Back to skill

Security audit

sos

Security checks for vulnerabilities and agentic risk

Overview

This skill gives careful instructions for filling a Word document from user-supplied spreadsheet data, with no hidden execution, persistence, or credential behavior found.

Before installing, confirm that a Chinese placeholder is acceptable for your documents or tell the agent to leave unsupported fields blank or use your preferred marker. Also note that the skill references a grounding-and-QA file that is not included in the package, so review outputs carefully for unsupported claims and formatting issues.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill hard-codes the Chinese placeholder 待确认 for unsupported fields without any user opt-in, document locale check, or surrounding language detection. In a document-generation workflow, this can introduce unintended foreign-language content into outputs, causing confusion, downstream processing errors, or inadvertent disclosure that the content was machine-filled or left unresolved.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.