ClawHub
Back to skill

Security audit

聘才猫（Pincaimao）在线面试

Security checks across malware telemetry and agentic risk

Overview

This interview skill’s external API use fits its purpose, but its privacy disclosure appears incomplete because sensitive interview reports can also be sent to a user-provided callback URL.

Install only if you are comfortable sending resumes, job descriptions, candidate answers, and interview reports to the Pincaimao service. Use callback URLs only for domains you control or explicitly trust, and avoid uploading unnecessary personal data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Intent-Code Divergence

Medium
Confidence
98% confidence
Finding
The documentation states the skill does not transmit data to any endpoint other than api.pincaimao.com, but the same file explicitly supports an arbitrary `url_callback` for report delivery. That contradiction can mislead users and reviewers about where sensitive interview data may be sent, weakening informed consent and security review.

External Transmission

Medium
Category
Data Exfiltration
Content
# Step 2: 第一轮 - 开始面试
QUERY=$(echo "$JOB_INFO" | cut -c1-20)
ROUND1=$(curl -s -X POST 'https://api.pincaimao.com/agents/v1/chat/chat-messages' \
  -H "Authorization: Bearer $PCM_ONLINE_INTERVIEW_KEY" \
  -H 'Content-Type: application/json' \
  -d "{
Confidence
90% confidence
Finding
This duplicate finding also reflects external transmission of interview initialization data to api.pincaimao.com. The skill context makes the transmission expected, but not risk-free, because job descriptions and attached candidate information may be confidential.

External Transmission

Medium
Category
Data Exfiltration
Content
# Step 3: 后续轮次 - 提交回答
USER_ANSWER="我有5年销售经验，擅长大客户开拓..."
curl -s -X POST 'https://api.pincaimao.com/agents/v1/chat/chat-messages' \
  -H "Authorization: Bearer $PCM_ONLINE_INTERVIEW_KEY" \
  -H 'Content-Type: application/json' \
  -d "{
Confidence
90% confidence
Finding
This duplicate finding reflects repeated transmission of user answers and conversation context to the external service. The multi-turn design increases exposure because sensitive data may accumulate over several rounds under one conversation.

External Transmission

Medium
Category
Data Exfiltration
Content
RESUME_FILE="/path/to/resume.docx"

# Step 1: 上传简历（可选）
UPLOAD=$(curl -s -X POST 'https://api.pincaimao.com/agents/v1/files/upload' \
  -H "Authorization: Bearer $PCM_ONLINE_INTERVIEW_KEY" \
  -F "file=@$RESUME_FILE")
COS_KEY=$(echo "$UPLOAD" | python3 -c "import sys,json; print(json.load(sys.stdin)['cos_key'])")
Confidence
89% confidence
Finding
Uploading a resume file to the external file API transmits highly sensitive personal information outside the local environment. In this skill context that behavior is intentional, but it still constitutes a real security/privacy risk because resume contents and returned storage keys may be exposed or mishandled.

External Transmission

Medium
Category
Data Exfiltration
Content
# Step 2: 第一轮 - 开始面试
QUERY=$(echo "$JOB_INFO" | cut -c1-20)
ROUND1=$(curl -s -X POST 'https://api.pincaimao.com/agents/v1/chat/chat-messages' \
  -H "Authorization: Bearer $PCM_ONLINE_INTERVIEW_KEY" \
  -H 'Content-Type: application/json' \
  -d "{
Confidence
90% confidence
Finding
This duplicate finding also reflects external transmission of interview initialization data to api.pincaimao.com. The skill context makes the transmission expected, but not risk-free, because job descriptions and attached candidate information may be confidential.

External Transmission

Medium
Category
Data Exfiltration
Content
# Step 3: 后续轮次 - 提交回答
USER_ANSWER="我有5年销售经验，擅长大客户开拓..."
curl -s -X POST 'https://api.pincaimao.com/agents/v1/chat/chat-messages' \
  -H "Authorization: Bearer $PCM_ONLINE_INTERVIEW_KEY" \
  -H 'Content-Type: application/json' \
  -d "{
Confidence
90% confidence
Finding
This duplicate finding reflects repeated transmission of user answers and conversation context to the external service. The multi-turn design increases exposure because sensitive data may accumulate over several rounds under one conversation.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.