ClawHub
Back to skill

Security audit

meta-boss-skills

Security checks for vulnerabilities and agentic risk

Overview

This skill is not clearly malicious, but it asks the agent to create local skill files and keep persistent learning records while handling potentially sensitive workplace material, without clear user control or retention limits.

Review this before installing if you plan to process real chats, email, meeting notes, or company documents. Use it only in a workspace where generated files under boss-related paths are acceptable, and avoid providing confidential material unless you are comfortable with derived profiles and learner notes being written locally.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
82% confidence
Finding
The skill describes behavior that writes generated artifacts into filesystem paths (`bosses/{slug}/`) but does not declare corresponding permissions. Undeclared write capability undermines least-privilege expectations and can surprise users or host systems with file modifications, especially in an agent context where skills may be trusted based on metadata.

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding
The declared purpose presents the skill as a meta-skill for distillation, verification, and orchestration, but the analyzed behavior includes persistent local logging of execution counts, failures, and notes. Hidden persistence is security-relevant because it can retain sensitive operational data beyond user expectations, creating privacy, data retention, and secondary exposure risks.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The workflow explicitly says to write generated bundles into `bosses/{slug}/` and generate skills via a writer script, but it does not warn the user that local files will be created or modified. In a skill ecosystem, silent filesystem changes are dangerous because they can overwrite data, create unexpected artifacts, or be abused if path values are influenced by untrusted input.

Static analysis

No suspicious patterns detected.