Security audit
Ai Tool Benchmark
Security checks for vulnerabilities and agentic risk
Overview
This is a simple benchmarking guide for comparing AI tools, with no executable code, persistence, credential use, or hidden data access found.
This skill appears safe to install from a security perspective. Treat the included tool rankings and pricing/performance claims as informational rather than authoritative, and verify the methodology and current pricing before using it for purchasing or tooling decisions.
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.