ClawHub
Back to skill

Security audit

Podcast Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently generates short podcast audio using web search and a disclosed XFYUN TTS dependency, with normal privacy and dependency risks to consider.

Install only if you trust the separate xfyun-tts skill and are comfortable using XFYUN API credentials. Avoid private, confidential, or regulated topics because prompts, retrieved facts, and generated script text may be processed by external services.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The skill instructs the agent to execute shell commands, search the local filesystem for another skill, and invoke a Python script directly. This creates an unnecessary command-execution and path-discovery surface for a workflow that should stay within constrained tool calls, and any variable interpolation in the generated script/content increases the risk of command injection, arbitrary file access, or execution of an unexpected script.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger phrases are broad enough to match ordinary conversational requests such as '播客' or '帮我做一段音频', which can cause the skill to activate unexpectedly. Unintended activation can lead to unsolicited web fetching, external service use, and file generation/transmission without clear user intent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill retrieves user-requested topic content from the web and then sends script text to an external TTS provider, but the description does not clearly warn users that their prompts and derived content will leave the local environment. This is a privacy and data-governance risk, especially if users request podcasts on sensitive, proprietary, or personal topics.

Static analysis

No suspicious patterns detected.