ClawHub
Back to skill

Security audit

Baoyu Comic

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed comic-generation workflow with expected local file output, preference storage, image-generation integration, and PDF assembly.

Install if you are comfortable with a comic tool that saves generated files, may store reusable preferences, invokes an image-generation skill, and runs bun or npx to create a PDF. Choose project-level preferences if you do not want defaults shared across projects.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
87% confidence
Finding
The workflow instructs the skill to inspect another installed skill's SKILL.md and infer its capabilities before proceeding. That expands the skill's access beyond comic creation into enumerating and reasoning about other local components, which can expose environment details and create cross-skill trust and prompt-injection risks if those files contain adversarial content. In this context, reading another skill may be operationally useful, but it still violates least privilege.

Context-Inappropriate Capability

Medium
Confidence
84% confidence
Finding
The workflow directly invokes an external script to merge outputs into a PDF. Script execution introduces a code-execution boundary inside a content-generation skill; if the script path, runtime variables, or comic directory are attacker-influenced, this can lead to unintended command execution or unsafe file handling. The merge step fits the feature set, but implementing it via shell/script invocation increases risk unnecessarily.

Session Persistence

Medium
Category
Rogue Agent
Content
│
        ▼
┌─────────────────────┐
│ Create EXTEND.md    │
└─────────────────────┘
        │
        ▼
Confidence
88% confidence
Finding
Create EXTEND.md │ └─────────────────────┘ │ ▼ Continue to Step 1 ``` ## Questions **Language**: Use user's input language or preferred language for all questions. Do not alwa

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.