ClawHub

Photo Video

Security checks across malware telemetry and agentic risk

Overview

This photo/video skill appears purpose-aligned, but it needs review because it can automatically connect and send broad user prompts or media to Nemo Video without a clear consent step.

Install only if you are comfortable sending selected photos/videos, prompts, and related session data to Nemo Video's cloud service. Before using it, require an explicit confirmation step before connecting, uploading files, or forwarding prompts, and avoid sensitive personal or confidential media unless you trust the service's retention and privacy practices.

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 (3)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill advertises broad trigger phrases such as "turn my photos or images" and encourages automatic setup on first interaction, which increases the chance of unintended invocation from ordinary conversation. Because invocation leads to authentication and connection to a third-party processing API, accidental triggering can cause unintended data flow and external actions before the user has clearly consented.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The routing table sends "Everything else" to the SSE chat/edit path, which is an overly broad catch-all with no exclusion criteria. In practice this can route unrelated user text to the remote backend, causing unintended third-party disclosure of prompts or edits and making accidental invocation more likely in a conversational environment.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill states that photos are processed through a cloud rendering pipeline and instructs the agent to connect automatically, but it does not present a clear user-facing warning that uploaded media, prompts, and session data are sent to a third-party service. This undermines informed consent and can expose sensitive images or text to an external processor without sufficiently explicit disclosure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal