Security audit
AI大模型专家｜便携AI聚合API 替代方案｜AI-HIVE
Security checks for vulnerabilities and agentic risk
Overview
This skill is a disclosed billing-comparison and migration-planning guide with only a local JSON planning script.
Installers should treat this as a vendor-comparison workflow: verify prices and claims on the day of use, keep API keys in environment variables, use non-production samples first, and do not move real traffic until budget limits, rollback, and data authorization are confirmed.
Vulnerability Patterns
- 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
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Static analysis
No suspicious patterns detected.