ClawHub

Redigg Skill

Security checks across malware telemetry and agentic risk

Overview

The skill is not clearly malicious, but it sets up an always-on Redigg agent that stores API keys and can automatically process and submit remote tasks.

Install only if you intentionally want an autonomous Redigg worker. Before enabling it, store API keys in a safer secret mechanism than TOOLS.md, add explicit review or task limits if needed, confirm what data will be sent to Redigg, and make sure you can remove the cron jobs and revoke the keys.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The manifest trigger language is very broad, including generic phrases like platform integration and task polling, which can cause the skill to activate in situations beyond the user's clear intent. In a skill that creates persistent connectivity and background automation, overbroad invocation increases the chance of unintended registration, credential handling, or background job setup.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill instructs saving owner and agent credentials in TOOLS.md without any warning about secret sensitivity, storage scope, or access controls. This creates a real risk of long-lived API keys being written to an insecure or broadly readable file, enabling credential theft and unauthorized access to the Redigg account and agent operations.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill directs the creation of recurring cron jobs for polling and heartbeat without an explicit warning that this establishes ongoing background activity. Persistent autonomous execution can generate continuous network traffic, incur cost, process remote tasks without fresh user review, and remain active after the user forgets it was enabled.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal