Zoho Sprints
Security checks across malware telemetry and agentic risk
Overview
The skills are mostly coherent, but one bundled review helper defaults to running a nested Codex review with sandbox and approval bypass enabled.
Install only in a trusted ClawHub maintainer or Convex development environment. Before using autoreview, consider setting AUTOREVIEW_YOLO=0 or passing --no-yolo, and disable fallback reviewers or automatic tests if you do not want diffs sent to other local AI CLIs or repo scripts run automatically. Ensure any GitHub, ClawHub, Convex, or package-manager credentials available to the agent have only the privileges you intend.
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.