Security audit
ia-verification-before-completion
Security checks across malware telemetry and agentic risk
Overview
This is a markdown-only engineering discipline skill that asks agents to verify work before claiming completion, with no hidden execution, credential use, or unrelated data access.
Before installing, understand that this skill will push the agent to run fresh verification commands before making completion claims. That may include tests, builds, local servers, git status checks, temporary worktrees, or migration dry runs depending on the project, so users should still review high-impact commands before allowing them.
SkillSpector
By NVIDIA
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
VirusTotal
64/64 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.