ClawHub
Back to skill

Security audit

OpenClaw P2P

Security checks across malware telemetry and agentic risk

Overview

This skill is meant for agent-to-agent messaging, but it needs review because it can send local data to external peers and runs unreviewed code outside the submitted skill files.

Install only if you trust the publisher and can verify the missing runtime implementation it executes. Do not send secrets, credentials, private files, customer data, or internal project context through this skill unless the peer and relay/privacy properties are understood and the transfer is explicitly approved.

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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill documents use of environment-derived paths and executable commands but declares no corresponding permissions or capability requirements. This can mislead users and policy systems about what the skill can access, reducing security transparency and increasing the chance that networked code is run with broader access than expected.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill encourages real-time messaging over public Nostr relays and local identity persistence without clearly warning that discovery metadata, encrypted messages, and transcripts may traverse third-party relays and be stored locally. In a P2P communication skill, this omission is especially dangerous because users may share sensitive prompts, data, or coordination details under the false impression that the channel is private or ephemeral.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The documented sendfile command normalizes transferring arbitrary file contents to another agent over the network but provides no warning about exfiltration risk, persistence, or trust boundaries. Given this skill’s purpose is inter-agent communication with unknown peers discovered via public relays, omitting that warning materially increases the chance of sensitive local data being disclosed.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.