Security audit
Evolve Self-Improving
Security checks for vulnerabilities and agentic risk
Overview
This skill stores carefully filtered local learning notes for future use, with no evidence of hidden network access, credential handling, or destructive behavior.
Before installing, be comfortable with the agent maintaining a local ~/.agent-knowledge/ memory store. The artifact includes reasonable controls for when to save, what not to save, and how to validate entries, but users should still avoid letting sensitive personal, customer, credential, or confidential data become memory entries.
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.