ClawHub

Video Searching

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed video-search integration that sends searches to Memories.ai using an API key, with no evidence of hidden, destructive, or unrelated behavior.

Install this only if you are comfortable sending video-search queries to Memories.ai under your MEMORIES_API_KEY. Avoid confidential or regulated search terms, use a scoped or dedicated API key if available, and expect a possible runner-path fix if the packaged root script is not found at the documented scripts/ path.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill explicitly instructs the agent to execute a shell script (`scripts/run_video_query.sh`) and requires `curl`/`jq`, but the manifest does not declare corresponding permissions. Undeclared shell capability weakens the trust boundary for users and platforms because the skill can invoke local commands and make networked API calls without transparent permission signaling.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill sends user queries to the external Memories.ai API, but the user-facing description does not clearly warn that prompts and possibly sensitive search terms leave the local environment. This creates a privacy and consent risk, especially if users provide proprietary, personal, or regulated information while expecting local-only processing.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal