Security audit
DeepThink — 深度思考模式
Security checks for vulnerabilities and agentic risk
Overview
This skill is a simple structured-thinking workflow with a local Python template script and no evidence of hidden data access, persistence, or network behavior.
Reasonable to install if you want a lightweight reasoning-template helper. Be aware it is an initial-release, low-trust publisher skill and its examples overstate the script's current CLI options, but the reviewed code only prints local templates and does not access private data or the network.
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.