ClawHub
Back to skill

Security audit

深度技术报告生成器

Security checks across malware telemetry and agentic risk

Overview

The skill openly generates technical reports, but its normal workflow can publish them to IMA and Tencent Docs using local credentials without a clear approval gate.

Install only if you want generated reports to be publishable to IMA and Tencent Docs. Before use, verify the target knowledge base, Tencent Docs account, sharing permissions, and IMA credentials, and review each report for confidential content before allowing any sync.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The skill advertises very broad trigger phrases such as generating reports, tutorials, onboarding docs, and explaining technical concepts, which can cause over-invocation outside the user's intended scope. In an agent environment, overly permissive routing increases the chance this skill is selected when users did not expect file generation and downstream publication behavior.

Missing User Warnings

High
Confidence
96% confidence
Finding
The description promises automatic syncing to IMA and Tencent Docs but does not prominently warn the user that generated content may be sent to external services. This creates a real risk of unintended data disclosure, especially if prompts or generated reports contain internal architecture details, proprietary information, or sensitive training material.

Missing User Warnings

High
Confidence
99% confidence
Finding
The workflow mandates automatic distribution to a specific IMA knowledge base and Tencent Docs without any confirmation gate, sensitivity review, or approval step. In context, this is more dangerous because the skill is designed for technical reports, which commonly include non-public implementation details, making silent exfiltration to third-party or shared systems a serious confidentiality risk.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The document instructs reading API credentials from local files and then transmitting them in request headers, but provides no warning about secret handling, storage hygiene, shell history/logging exposure, or scope restrictions. In a reusable agent skill, this can normalize unsafe credential practices and lead users or downstream automation to expose long-lived secrets to external services without adequate safeguards.

External Transmission

Medium
Category
Data Exfiltration
Content
API_KEY=$(cat ~/.config/ima/api_key)

# Step 2: Preflight Check（验证连接）
curl -s "https://api.ima.qq.com/open/knowledge/check" \
  -H "client_id: $CLIENT_ID" \
  -H "api_key: $API_KEY"
Confidence
88% confidence
Finding
curl -s "https://api.ima.qq.com/open/knowledge/check" \ -H "client_id: $CLIENT_ID" \ -H "api_key: $API_KEY" # Step 3: 检查重名（避免重复上传） curl -s "https://api.ima.qq.com/open/knowledge/check_repeated_na

External Transmission

Medium
Category
Data Exfiltration
Content
API_KEY=$(cat ~/.config/ima/api_key)

# Step 2: Preflight Check（验证连接）
curl -s "https://api.ima.qq.com/open/knowledge/check" \
  -H "client_id: $CLIENT_ID" \
  -H "api_key: $API_KEY"
Confidence
88% confidence
Finding
https://api.ima.qq.com/

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.