ClawHub
Back to skill

Security audit

上市策略引擎(免费版)

Security checks for vulnerabilities and agentic risk

Overview

This go-to-market planning skill is mostly a strategy template, but it asks for command execution and file-writing capability without explaining why those powers are needed.

Install only if you are comfortable granting a planning template broader agent capabilities than it appears to need. Prefer using it in a constrained workspace, avoid providing sensitive business data to callback URLs unless you trust the endpoint, and do not let it run shell commands unless you explicitly requested and reviewed them.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill is presented as a go-to-market planning/documentation aid, but it declares `exec` capability without any clear operational need, workflow justification, or guardrails. That expands the attack surface unnecessarily: a prompt injection or ambiguous user request could cause shell command execution in an environment where only text analysis should be needed.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill accepts a `callback_url` for asynchronous notification but does not warn users that this may transmit output or metadata to an external destination. In an agent environment, that can enable unintended data exfiltration, SSRF-style misuse, or disclosure of sensitive business inputs if users supply untrusted callback endpoints.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
Listing `exec` and `write` capabilities without a user-facing warning hides that the skill may alter files or execute system commands. Because the documented function is strategic planning rather than system administration, these capabilities are contextually suspicious and increase the chance of unsafe actions being taken without informed consent.

Static analysis

No suspicious patterns detected.