ClawHub
Back to skill

Security audit

picoads

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent paid ad-network integration, but it gives an agent authority to create USDC-linked ad marketplace actions without enough built-in approval or spending guardrails.

Review before installing. Use this only if you want the agent to participate in a paid ad marketplace, and set strict rules requiring your explicit approval for registration, bids, asks, delivery submissions, confirmations, disputes, budgets, and any audience-facing sponsored content.

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
Findings (2)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly describes actions that create bids, asks, deliveries, disputes, and settlements tied to USDC payments, but it does not provide clear user-facing warnings that these are real financial and wallet-impacting operations. In an agent context, this omission is dangerous because a model may present these operations as routine API calls and initiate or recommend irreversible monetary actions without informed user consent.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The registration flow states that obtaining an API key requires a '$1 USDC via x402 payment' but does not prominently warn that registration triggers a real charge. This is risky because users or downstream agents may interpret registration as a normal setup step rather than a paid action, leading to unintended spending.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.