ClawHub
Back to skill

Security audit

Linear Autopilot

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Linear automation tutorial, but it asks users to connect credentials, auto-process tasks, update Linear, send Discord messages, and push git changes with limited safeguards.

Review before installing. Use least-privilege Linear and Discord tokens, avoid plaintext secrets where possible, keep Discord task channels restricted, require explicit bot mentions unless you really want unattended processing, disable autoPush until tested, and verify any referenced scripts before running them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill presents a reassuring security note claiming it does not collect or transmit credentials, yet the workflow explicitly instructs users to configure API keys, bot tokens, webhook URLs, and git-enabled automation through third-party services. That mismatch can create a false sense of safety and encourages users to place sensitive credentials into an automation chain involving external platforms, increasing the chance of credential leakage, misuse, or over-broad trust.

Vague Triggers

Medium
Confidence
79% confidence
Finding
The trigger keywords are broad terms such as 'processing', 'autopilot', 'task', and 'automate', which can cause the skill to activate in contexts far beyond the intended Linear workflow. In a skill with exec capability and instructions for automated task execution, over-broad activation increases the risk of unintended invocation, accidental command execution, and user confusion about when the automation should run.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.