ClawHub
Back to skill

Security audit

prd-generator-from-prototype

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only PRD generation skill with no code execution, dependencies, persistence, or hidden data access.

Install only if you want a Chinese-language PRD template workflow. Review generated implementation notes, API details, and estimates before treating them as authoritative, especially when the prototype or scope is incomplete.

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 (5)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill claims it will not expand requirement scope, yet later permits optional appendices with development advice, effort estimates, and risk prompts that are not necessarily derivable from the provided prototype or declared scope. This internal contradiction weakens containment guarantees and can be exploited to elicit extra content outside approved boundaries, undermining user intent and potentially introducing speculative requirements.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill is presented as a PRD generator constrained by prototype and scope, but these sections explicitly allow output of implementation-oriented artifacts such as technical implementation notes, API definitions, DDL-style field tables, UI design appendices, and development suggestions. That broadens the agent from documentation synthesis into solution design, which can cause it to invent architecture, interfaces, or data structures not grounded in the user-provided prototype and thereby produce unauthorized or unsafe downstream guidance.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The file requires output to use Chinese-numbered headings and specific Chinese formal phrasing, which imposes a language/locale constraint on all generated PRDs. Because the skill does not offer an opt-in choice or explain a justified region-specific requirement, this is a natural-language policy violation under the locale-policy rule.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The manifest description is written entirely in Chinese and describes the skill behavior in a way that implies a fixed language/locale, without offering the user any choice or opt-in. This can violate language/locale policy when the skill is used in broader contexts where users may expect language flexibility.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The section explicitly requires heading and document structure formats using Chinese numerals and Chinese punctuation as a mandatory rule. This imposes a specific language/locale convention on all outputs without offering user opt-in or explaining a justified region-specific constraint, which matches the natural-language locale policy violation category.

Content

No source excerpt is available for this finding.