ClawHub
Back to skill

Security audit

商机挖掘

Security checks for vulnerabilities and agentic risk

Overview

This skill is coherent for Cue business research, but it asks the agent to fetch and run unpinned external code that uses a local Cue API key.

Install only if you trust the Cue runner repository and are comfortable letting it use your local Cue API key and credits. Prefer reviewing or pinning the runner code before use, and confirm each paid research run carefully.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill instructs the agent to `git pull` or clone and then execute tooling from external repositories at runtime. That expands the skill from business research into arbitrary supply-chain code retrieval and execution, creating a clear path for repository compromise, mirror tampering, or unexpected code changes to affect the host environment.

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
The skill explicitly relies on a local API key stored in `~/.cue/config.json` and directs use of a runner that automatically reads it. Accessing local secrets is sensitive and only loosely justified in the document, and it risks exposing or misusing user credentials through the runner or downstream services.

Static analysis

No suspicious patterns detected.