ClawHub
Back to skill

Security audit

Tick-Coord: Multi-Agent Task Coordination

Security checks across malware telemetry and agentic risk

Overview

This task coordination skill appears legitimate, but it can expose project task data through a network dashboard and remote sync that are not clearly disclosed up front.

Install only if you are comfortable with agents editing TICK.md task records and potentially syncing them through Git or Convex. Prefer localhost-only dashboard use, stop the background server after inspection, avoid running it on public or shared networks, and review task contents for sensitive project details before any Git push or Convex sync.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill is presented as a local Git-backed Markdown coordination tool, but later introduces a networked dashboard and remote Convex sync path. That mismatch can mislead users and downstream policy systems about data flow and trust boundaries, increasing the chance that project/task data is exposed off-machine without informed consent.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The documentation instructs starting a detached development server bound to 0.0.0.0, making it reachable from other hosts on the network. Even with a warning, exposing an unauthenticated dev server increases the attack surface and can leak sensitive task/project data to anyone with network access, especially because the process is designed to persist in the background.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The open-ended trigger phrase 'or similar' creates ambiguous activation boundaries for launching the dashboard. In an agent setting, vague triggers can cause unintended execution of server-starting behavior in response to loosely related user requests, leading to unnecessary network exposure or background process creation without clear user intent.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs syncing task data to a remote Convex deployment but does not place a clear, user-facing warning immediately at that step that project/task contents will leave the local machine. Because TICK data may contain sensitive operational details, this can result in accidental exfiltration to third-party infrastructure under the guise of normal coordination.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.