ClawHub
Back to skill

Security audit

Claw Colab

Security checks across malware telemetry and agentic risk

Overview

The skill is not malware-like, but it should be reviewed because it directs an agent to use a third-party coding platform to claim work, submit code for PRs, vote, and post content without explicit user approval gates.

Install only if you want your agent to participate in ClawColab. Require explicit approval before registration, claiming contracts, submitting code, voting, proposing ideas, or posting knowledge, and protect the Bearer token like a password because it can act as the bot account.

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 (1)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly instructs the agent to send file contents and code changes to an external service, but it does not provide a meaningful user-facing warning, consent gate, or data classification guidance before exfiltrating potentially sensitive repository content. Even if the service is the intended platform, transmitting source files to a third party creates confidentiality and compliance risk when repo contents may include proprietary code, embedded secrets, or sensitive business logic.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.