ClawHub
Back to skill

Security audit

企业尽调-分析生成

Security checks for vulnerabilities and agentic risk

Overview

This skill performs disclosed company due-diligence research and document generation, with sensitive inputs that users should share carefully.

Use this skill when you intend the agent to research a company online and generate local Word reports. Provide only necessary documents, redact irrelevant personal or confidential details, and confirm any API tokens or MCP connectors before enabling them.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
97% confidence
Finding
The skill invokes Python scripts, reads environment variables like QICHACHA_TOKEN, performs network access to external data sources, and generates .docx files, yet declares no permissions. This creates a mismatch between stated and actual capabilities, undermining sandboxing and user consent expectations and potentially allowing sensitive file, network, or environment access without explicit review.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README explicitly encourages users to upload sensitive materials such as financial statements, contracts, and business registration documents, but it provides no handling restrictions, minimization guidance, redaction advice, or retention/confidentiality warnings. In a due-diligence workflow, these materials commonly contain non-public financial, legal, and personal data, so the omission increases the risk of oversharing sensitive information into the agent pipeline.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The description uses broad trigger phrases such as '帮我做尽调', '企业背景调查', and 'due diligence', which can overlap with routine research or generic company questions. Over-broad activation can cause the wrong skill to run, leading to unnecessary external lookups, document processing, and report generation on sensitive user inputs.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger list includes ambiguous short phrases like 'DD' and broad phrases like '查一下这家公司' without scope constraints. In context, this skill can fetch public records, process user documents, and create files, so accidental invocation expands data handling beyond what a user may have intended.

Static analysis

No suspicious patterns detected.