Security audit
ai-architecture-harness-zh
Security checks for vulnerabilities and agentic risk
Overview
This is a documentation-only architecture guardrail skill with no executable payload, credential access, network behavior, or hidden persistence.
Review any documentation, test, or lint changes the agent proposes as normal repository changes. The skill is meant to strengthen architecture guardrails and shows no evidence of hidden execution, unrelated data access, or credential use.
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.