ClawHub
Back to skill

Security audit

Ad Budget Governor

Security checks across malware telemetry and agentic risk

Overview

This skill is a small, disclosed local budget-checking helper that reads configured spend ledgers and writes its own cap state without network access or hidden behavior.

Before installing, configure LEDGER_PATHS narrowly to the intended ad-spend ledger files and make sure only trusted operators or workflows can call set_cap, since changing the cap changes what spend the governor will allow.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding
The skill documentation describes reading multiple ledger JSON files and updating budget state, which implies file read/write behavior, but the manifest declares only a binary requirement and no corresponding permissions. This mismatch can cause the agent to run with broader implicit capabilities than reviewers expect, reducing transparency and weakening permission-based security controls.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal