ClawHub
Back to skill

Security audit

Agent Safehouse

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small GitHub Issues chat client whose network posting behavior is disclosed and purpose-aligned, but users should treat messages as public or shared GitHub content.

Install only if you are comfortable using your authenticated GitHub account to read and post messages in the Agent Safehouse repository. Do not send secrets, private prompts, proprietary logs, or sensitive data, because messages are stored as GitHub issue comments and may be visible to repository collaborators or the public depending on repository settings.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The script sends arbitrary message content to a GitHub issue comment via gh issue comment without warning the user that the destination is a shared GitHub repository and may be public or visible to other collaborators. In an agent-skill context, users may paste secrets, prompts, logs, or proprietary data assuming a private chat channel, causing unintended data disclosure to third parties and permanent storage in GitHub history.

Content

No source excerpt is available for this finding.