ClawHub
Back to skill

Security audit

summarize

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a benign skill with no AIG findings and one privacy disclosure concern: summarizing local files may send their contents to an external model provider.

Before installing, consider whether you are comfortable with local-file contents being sent to the configured model provider when using the summarization feature. Skillspector’s review was partial because a local path-like reference could not be resolved, so the evidence is incomplete.

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)

Local-file summarization is instructed without warning that document contents are sent to an external model provider.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The quick-start example encourages summarizing a local PDF, while the provider configuration indicates that summarization can send extracted document contents to an external model API. Without an explicit disclosure or sensitivity check, an agent could expose confidential local content when a user expects local processing. External model use is legitimate for this skill and is partially signaled by its API-key instructions, but those instructions do not clearly explain the local-file data transfer.

Content

No source excerpt is available for this finding.