ClawHub
Back to skill

Security audit

referral-coordination-desk

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple referral-coordination helper that only describes two user-directed workflows and does not include code, installation steps, persistence, or hidden access.

This skill appears safe to install for referral digest and packet creation workflows. Referral information can be sensitive, so users should only provide referral data they intend the agent to process and should keep use limited to the two supported stages.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (1)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill says to 'Use this skill for routine care coordination work' before later narrowing behavior to two stages, which creates an overly broad trigger surface. In an agent setting, that ambiguity can cause the skill to be invoked for adjacent care-coordination tasks outside the defined tool interfaces, increasing the chance of unintended tool use, data shaping, or workflow execution with sensitive referral data.

Static analysis

No suspicious patterns detected.