ClawHub

OADP Emit

Security checks across malware telemetry and agentic risk

Overview

This skill is documentation-only and mostly matches its discovery purpose, but it asks users to persistently advertise their agent and send identifying metadata to a third-party hub without enough privacy, scope, or rollback guidance.

Install only if you intentionally want this agent or web root to be discoverable. Before running the snippets, verify the onlyflies.buzz hub, use only public-safe agent metadata, avoid internal/private workspaces, back up files before changing them, and keep removal steps for AGENTS.md, .well-known/agent-protocol.json, robots.txt, and any HTML meta tags.

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 (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs the user to append a hidden OADP marker into a workspace file read on boot, modifying local state in a way that affects agent behavior and discoverability without any warning, consent, or rollback guidance. Because the marker is specifically designed for scanners to detect and influence future interactions, this creates an unnecessary persistence and trust-signal injection risk in the local workspace.

Missing User Warnings

High
Confidence
98% confidence
Finding
The skill directs users to transmit agent identity and registration data to a third-party external hub without any privacy notice, trust validation, data handling explanation, or consent flow. In the context of an agent skill, this is particularly dangerous because it can disclose the existence, identity, capabilities, and network-reachable metadata of the agent to an untrusted service, enabling tracking, targeting, or downstream abuse.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs users to create or overwrite public web-discoverability files such as .well-known metadata and robots.txt entries, which can alter site behavior, expose infrastructure metadata, and affect how automated systems discover the service. Doing so without warnings about overwrite risk, public exposure, or operational impact can unintentionally publish sensitive hub relationships or disrupt existing site configuration.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal