ClawHub
Back to skill

Security audit

Product Dev Workflow V5

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Markdown-only product development report template with no executable code, though it has rigid trigger and Chinese-only output rules users should understand.

Install this if you want a Chinese-language, one-shot product development report workflow. Be aware that it may trigger broadly for product-planning requests and may not naturally match non-Chinese language preferences without modification.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill’s trigger is excessively broad and uses mandatory language ('务必使用此skill'), which can cause the agent to invoke it for many common product-related prompts without sufficient intent disambiguation. This creates prompt-routing risk: users may be locked into a rigid, high-scope workflow that overrides user preferences, increases token usage, and can suppress safer or more appropriate handling paths.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

Forcing all output to Chinese regardless of user language preference can cause misalignment with user intent and reduce transparency if the user cannot read the response. While not a direct code-execution or data-exfiltration issue, it is a control and usability weakness that can degrade safe interaction by preventing clear review, consent, or correction of generated technical content.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The output quality section repeats a mandatory Chinese-only requirement, reinforcing language coercion across all responses. In context, this makes the behavior more persistent and harder for the agent to override, increasing the chance that users receive unusable or misleading output if they expected another language.

Content

No source excerpt is available for this finding.