ClawHub
Back to skill

Security audit

Client Proposal Generator

Security checks for vulnerabilities and agentic risk

Overview

This is a self-contained prompt-template skill for writing client proposals, with no code execution, credential access, persistence, or hidden data handling found.

Before installing, be aware that the README advertises an optional external done-for-you service where you would send project notes to the author. Using the local skill itself appears to only use the proposal details you choose to provide in prompts; avoid sharing confidential client information with any outside service unless you are comfortable with that arrangement.

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 (1)

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The file presents itself as a skill for generating client proposals via reusable prompts, but these lines introduce a manual 'Done-For-You' service where the author asks users to send project notes for direct delivery of a finished proposal. That contradicts the documented nature of the artifact as a self-contained skill and blurs whether the package's intent is software assistance or off-platform human service fulfillment.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.