ClawHub
Back to skill

Security audit

Persona Team Lead

Security checks for vulnerabilities and agentic risk

Overview

This team-lead skill fits its purpose, but it can post team information to Chat and append shared Sheet data without explicit confirmation or destination checks.

Install only if you are comfortable with an agent using Google Workspace tools for team coordination. Before posting or appending anything, verify the Chat space or Sheet, review the exact content, and use sanitization for sensitive team data.

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 (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The skill description and persona framing are broad enough to be invoked in many ordinary team-management contexts without clear boundaries on when the persona should activate. In an agent environment with tool access to calendar, email, chat, drive, and sheets, this can cause overbroad delegation and unintended execution of sensitive coordination actions from vague prompts.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The instructions explicitly encourage sharing standup outputs in Chat and appending OKR or other team data into shared Sheets, but they do not require a warning, confirmation, or recipient validation before exposing or modifying shared data. Because this persona is designed to aggregate team information and has access to collaboration tools, a vague or mistaken invocation could spread sensitive team details broadly or write incorrect data into shared systems.

Static analysis

No suspicious patterns detected.