ClawHub

Smart Spawn API

Security checks across malware telemetry and agentic risk

Overview

This instruction-only skill openly uses an external model-selection API, but it broadly sends task descriptions to that service and can influence sub-agent spawning without clear privacy or control limits.

Install only if you are comfortable sharing task descriptions with ss.deeflect.com. Avoid sending secrets, customer data, proprietary plans, credentials, or regulated information; review any decompose or swarm output before spawning agents, and set your own limits for budget, model allowlists, and number of spawned sessions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill is framed as applicable to essentially any task and encourages routing arbitrary work through an external model-selection API without clear activation boundaries or safety constraints. This can cause the agent to send sensitive or inappropriate task descriptions to a third party and overuse the skill in contexts where external disclosure or untrusted model selection is not acceptable.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill instructs sending the user's task description to ss.deeflect.com over HTTP requests but does not warn that prompts, project details, or potentially sensitive data leave the local agent environment. Because the advertised usage is broad ('any task'), users or downstream agents may unknowingly transmit confidential business logic, credentials, customer data, or proprietary plans to an external service.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal