ClawHub Publishing Workflow
Security checks across malware telemetry and agentic risk
Overview
The reviewed skills are repo-maintenance helpers with some powerful commands, but their use is disclosed, scoped, and tied to the stated ClawHub/Convex workflows.
Install only if you are comfortable giving an agent maintainer-level access in a ClawHub/Convex repo. Pay particular attention to the moderation and autoreview workflows: they can affect real ClawHub accounts/content or run nested review tooling with broad local authority, so use them with the stated confirmation and target-scoping rules.
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
63/63 vendors flagged this skill as clean.