ClawHub
Back to skill

Security audit

Prompt Optimizer Toolkit

Security checks for vulnerabilities and agentic risk

Overview

This is a local prompt-improvement toolkit whose only meaningful risk is that its optional prompt library can save prompts as plain local JSON files.

Install only if you are comfortable with optional saved prompts being written locally as plain JSON. Avoid saving API keys, credentials, proprietary prompts, or personal data unless you control the chosen storage path and cleanup/retention process.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
90% confidence
Finding
The manifest declares Python execution and the documented API includes a local prompt library with save functionality, implying file read/write capability, but no corresponding permissions or activation boundaries are declared. This creates a transparency and governance gap: users or the platform may not understand that the skill can persist or access local data, which increases the risk of unintended data exposure or unauthorized file modification.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The description advertises broad prompt analysis, optimization, templating, and library management capabilities without clear limits on when file-backed operations occur or what content the skill will store/process. Overly broad capability language can mislead users and downstream agents about the trust boundary, making it easier for sensitive prompts or workspace content to be handled in ways the user did not intend.

Missing User Warnings

Low
Confidence
81% confidence
Finding
The PromptLibrary persists arbitrary user-provided prompts to disk without any warning, consent flow, sensitivity check, or access control. Because prompts often contain secrets, proprietary instructions, or personal data, silent persistence can create unintended local data exposure and retention risk on shared systems or developer workstations.

Static analysis

No suspicious patterns detected.