ClawHub
Back to skill

Security audit

Seisoai

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent Seisoai media-generation gateway, but it needs Review because it can send user media to a third party, spend through x402, and invoke face-swap or voice-clone tools without enough explicit user-control and consent guidance.

Install only if you are comfortable letting the agent send prompts, uploaded or linked media, API credentials, and payment requests to Seisoai. Require explicit approval before paid x402 calls, check pricing first, avoid secrets or sensitive personal data in prompts and URLs, and use face-swap or voice-clone workflows only with clear authorization from the person involved.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
Findings (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill directs agents to send prompts and media to a third-party service but does not warn users that submitted text, images, audio, or video may be sensitive and will leave the local trust boundary. This creates a meaningful privacy and compliance risk because agents may forward confidential, personal, or regulated content to external endpoints without explicit user awareness or consent.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill exposes high-impact capabilities including face swap and voice cloning without any user-facing safety warning, consent requirement, or abuse guardrail in the documentation. These capabilities increase the risk of impersonation, non-consensual media generation, fraud, and reputational harm if an agent invokes them casually or on unverified user inputs.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.