ClawHub
Back to skill

Security audit

Onboarding Flow Designer(2)

Security checks for vulnerabilities and agentic risk

Overview

This is a non-executable UX planning skill for onboarding flows, with a privacy caveat around suggested push/email reactivation tactics.

Safe to install as a planning aid. Before using its generated onboarding plans in a real product, add explicit opt-in, unsubscribe/preferences, frequency limits, and privacy-law review for any push or email reactivation recommendations.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger phrases include very generic requests such as '设计新手引导' and '新用户怎么上手', which are common in ordinary product or UX discussion and could cause the skill to activate unintentionally. Unintended invocation can hijack unrelated conversations, leading to prompt/behavior collision and unexpected output generation in contexts where the user did not explicitly request this skill.

Natural-Language Policy Violations

Medium
Confidence
83% confidence
Finding
The skill title and content are entirely Chinese-oriented, and the finding indicates language output is forced without checking user preference. While not a classic security flaw, this can override user intent, reduce transparency, and contribute to unsafe agent behavior by unexpectedly changing communication mode during invocation.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly suggests re-engagement via push notifications without any guardrails around lawful basis, opt-in consent, frequency limits, or jurisdiction-specific privacy requirements. This can lead downstream agents or users to design onboarding flows that encourage unsolicited messaging practices, creating privacy, compliance, and user-trust risks.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The final output template normalizes push and email reactivation as standard recommendations but omits any warning about consent, preference management, or personal-data use. Because this is part of the required output structure, it increases the chance that generated plans will systematically include privacy-impacting outreach without appropriate safeguards.

Static analysis

No suspicious patterns detected.