Back to skill

Security audit

PRD Iterative

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Chinese-language PRD drafting workflow that asks for confirmation before using models, fetching linked material, or writing output files.

Install this if you want a Chinese PRD workflow that can run several writer/reviewer rounds. Confirm the models, output folder, round count, and any links carefully, and avoid providing private or sensitive source material unless you are comfortable with the agent processing it.

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
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases are broad enough to activate on ordinary requests for general PRD assistance, which can cause the agent to enter a more powerful workflow than the user intended. That raises the risk of unnecessary multi-agent execution, model selection prompts, file-path handling, and external content fetching without clear user consent for this specific orchestration.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

L021 explicitly states that model and delivery path must be user-confirmed before starting and that defaults cannot be used directly. However, L060-L063 and the standard configuration at L047-L052 present concrete default model and path values to use during startup, creating a direct contradiction in the documented operating intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The applicable scenarios section lists broad examples like '帮我写一个xxx的PRD' without boundaries, making accidental invocation likely during normal drafting help. In this skill's context, accidental activation is more dangerous because the workflow includes iterative delegation, path collection, and potential link retrieval, expanding the action surface beyond a simple writing request.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The description and origin fields are written entirely in Chinese and present the skill as a Chinese-language PRD workflow, but the manifest does not indicate that this locale restriction is optional or limited to a region-specific use case. Under the policy, language constraints should either be opt-in or clearly justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The entire PRD template is written as a fixed Chinese-language structure, with headings and guidance only in Chinese. Because the file does not state that it is intended specifically for Chinese-speaking users or offer an alternative language choice, it can be interpreted as imposing a locale/language preference without user opt-in.

Content

No source excerpt is available for this finding.