ClawHub
Back to skill

Security audit

AI KEY·这个月钱去哪了

Security checks for vulnerabilities and agentic risk

Overview

The skill’s revenue analysis fits its purpose, but its configured memory can retain confidential business details.

Provide the missing configuration and review the shared interaction instructions before use. Check where memory is stored and who can access it. If customer names, financial values, or payment patterns should not persist, instruct the agent to omit them from memory and remove existing notes. Retention controls were not supplied for review.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 skill persists sensitive business information without warning the user about retention.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The skill instructs the agent to persist revenue-attribution conclusions, dated financial values, accounting corrections, and recurring customer payment patterns without explaining retention or obtaining permission. Because this workflow analyzes identifiable customers and sensitive business records, those notes could expose confidential financial relationships through later memory access, reuse, or unintended sharing. The configured memory destination and legitimate calibration purpose reduce suspicion of malicious intent, but do not establish access controls or a retention policy; no external exfiltration is shown.

Content

No source excerpt is available for this finding.