ClawHub
Back to skill

Security audit

github

Security checks for vulnerabilities and agentic risk

Overview

The verdict is suspicious because scanners report an instruction to publish a runtime session URL in pull request bodies without warning the user or checking whether sharing is appropriate.

Consider the session URL disclosure risk before installing. The scan found one medium-severity issue, but its analysis was partial because a referenced local path could not be matched to a bundled artifact, so the available 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)

The skill instructs the agent to publish a runtime session URL in PR bodies without warning the user about sharing it.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The skill directs the agent to publish a runtime session URL in PR bodies without checking whether the session is intended for sharing or obtaining explicit authorization. PR readers could receive sensitive session metadata or access to conversation content if the linked session is accessible to them; requiring an exact runtime-supplied URL prevents fabrication but does not establish that disclosure is safe.

Content

No source excerpt is available for this finding.