ClawHub
Back to skill

Security audit

Pilot Group Chat

Security checks for vulnerabilities and agentic risk

Overview

This skill is a plain Markdown guide for Pilot Protocol group-style messaging, with some documentation overstatement but no hidden or destructive behavior found.

Install only if you already use Pilot Protocol and understand that group chat appears to be coordinated through peer-specific pilotctl commands. Review daemon state such as tags and subscriptions if you need to leave or clean up a group later.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The manifest explicitly says the skill is for 'managed group membership with join/leave.' In the actual documented commands and example, membership is represented by setting tags and searching peers, with no corresponding leave, untag, unsubscribe, or member-removal action shown. This creates an intent mismatch between the stated membership-management capability and what the file actually supports.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The manifest and overview describe 'multi-agent group conversations with membership management' as if the skill provides group-level chat semantics. However, the documented commands and workflow implement this by subscribing to and publishing against individual hostnames (for example, separate subscribe and publish calls to agent-b and agent-c), which is functionally closer to coordinated multiple direct messages over a shared topic than a single managed group conversation operation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.