ClawHub
Back to skill

Security audit

Scientific Figure Analysis Pipeline

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent scientific figure analysis workflow, but users should know it can upload extracted paper figures to Moonshot/Kimi for vision analysis.

Install only if you are comfortable sending selected extracted figures and prompts to Moonshot/Kimi. Do not use it on unpublished, confidential, regulated, patient, or proprietary PDFs unless you have permission, and keep the API key in an environment variable or secret manager rather than hardcoding it.

SkillSpector

By NVIDIA
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 (5)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The README explicitly describes sending extracted scientific figure images to a third-party API for analysis, but it does not warn users that document-derived content may leave the local environment. In this skill’s context, uploaded figures may contain unpublished research, proprietary data, patient-related imagery, or other sensitive material, so omission of a clear data-transfer/privacy notice creates a real security and compliance risk.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill explicitly instructs sending extracted PDF images and nearby text to a third-party vision API, but it does not clearly require user consent, data minimization, or a warning that unpublished papers, copyrighted figures, or sensitive research content may leave the local environment. In this context, the risk is elevated because the workflow is designed to extract and forward scientific figures and text at scale, which can expose confidential or proprietary material.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The document instructs users to send scientific figure images and prompts to a third-party vision API, but it does not warn about privacy, confidentiality, licensing, or data retention implications. In a literature-mining workflow, figures may come from unpublished, proprietary, or access-restricted papers, so silent external transmission can expose sensitive research material.

External Transmission

Medium
Category
Data Exfiltration
Content
| 属性 | 值 |
|------|-----|
| **模型** | `kimi-k2.6` |
| **API Base** | `https://api.moonshot.cn/v1` |
| **兼容性** | 完全兼容 OpenAI SDK |
| **视觉能力** | 图片 + 视频 |
| **上下文** | 256K tokens |
Confidence
94% confidence
Finding
https://api.moonshot.cn/

External Transmission

Medium
Category
Data Exfiltration
Content
client = OpenAI(
    api_key=os.environ.get("MOONSHOT_API_KEY"),  # 或直接传入字符串
    base_url="https://api.moonshot.cn/v1",
)

# 1. 读取图片并 base64 编码
Confidence
95% confidence
Finding
https://api.moonshot.cn/

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
references/kimi-k2.6-vision-api.md:94