Writing Skills
Security checks across malware telemetry and agentic risk
Overview
This skill is a coherent guide for writing and testing OpenClaw skills, with expected notes around persistent skill edits and an optional Graphviz helper script.
Reasonable to install if you want help authoring skills. Review any generated or edited skill files before keeping them because they can steer future agents, keep pressure-testing prompts confined to controlled testing, and only run render-graphs.js on skill directories and Graphviz installations you trust.
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
66/66 vendors flagged this skill as clean.