ClawHub
Back to skill

Security audit

pollinations-sketch-note

Security checks across malware telemetry and agentic risk

Overview

This skill appears to do what it claims: create sketch-note images using external search and image-generation services, with privacy and credential-handling caveats.

Install only if you are comfortable sending image topics/prompts to external services and storing provider API keys in environment variables. Avoid sensitive or confidential topics, consider specifying an explicit output path, and be aware that the Pollinations key is placed in the request URL.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Tainted flow: 'params' from os.environ.get (line 253, credential/environment) → requests.get (network output)

Critical
Category
Data Flow
Content
}

    # print(f"🎨 正在生成背景图（风格：{style[:50]}...）")  # 静默模式
    response = requests.get(url, params=params, timeout=60)

    if response.status_code == 200:
        with open(output_path, 'wb') as f:
Confidence
96% confidence
Finding
response = requests.get(url, params=params, timeout=60)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The invocation phrases are very broad (e.g., generic requests to generate an image or knowledge card) and can easily match normal user conversation, causing the skill to trigger unexpectedly. In an agent setting, this can lead to unintended external searches and image-generation requests, with privacy, cost, and user-consent implications.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The README states that the skill automatically searches Wikipedia/Baidu Baike and generates content, but it does not clearly warn users that their topic queries are sent to external services. This creates a transparency and privacy risk because users may unknowingly disclose sensitive or proprietary topics to third-party APIs.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger phrases are generic enough to match ordinary image-generation requests, which can cause this skill to be invoked when a user did not specifically intend to use its auto-search and summarization workflow. That increases the chance of unintended data disclosure to external APIs and unexpected execution of network-backed actions.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The manifest states that the skill automatically searches topics and summarizes them, but it does not disclose that user-supplied topics may be transmitted to third-party services via API keys such as POLLINATIONS_API_KEY and TAVILY_API_KEY. Users may unknowingly submit sensitive subjects to external providers, creating privacy and data-handling risk.

Missing User Warnings

Medium
Confidence
82% confidence
Finding
The script transmits user-provided themes to third-party services and also fetches remote content without clearly warning the user in help text or documentation. In an agent-skill context, undisclosed external transmission is more concerning because users may assume local-only processing while their prompts are sent off-host.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger phrases are very broad and overlap with ordinary image-generation requests, so the skill may be invoked when a user did not specifically intend to use this tool. In this skill's context, unintended invocation is more concerning because it can cause external Tavily/Pollinations API calls and file creation, expanding the effect of a simple conversational request beyond local text generation.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill description does not clearly warn that it sends user topics to external APIs and writes PNG files, which undermines informed consent and can expose user-provided content to third parties. In this context, the risk is amplified because the skill automatically searches, summarizes, generates an image remotely, and saves output, all of which have privacy and side-effect implications.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.