Security audit
Optimize Prompt
Security checks for vulnerabilities and agentic risk
Overview
This skill coherently optimizes prompts without executing the underlying task, and I found no hidden persistence, credential access, destructive behavior, or unrelated data flows.
Installers should understand that this skill rewrites and audits prompts, not the task itself. Review the optimized prompt before sending it to another agent, and avoid including secrets or private data if your configured model adapter sends prompts to an external provider.
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.