OpenClaw Knowledge Runtime
Security checks across malware telemetry and agentic risk
Overview
This skill is a coherent knowledge-memory design aid with disclosed local persistence and no evidence of hidden execution, credential access, or network behavior.
Before installing, decide which memory files this skill may read and write, and avoid storing secrets, raw logs, full prompt transcripts, or speculative notes in shared or published knowledge stores.
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
65/65 vendors flagged this skill as clean.