xy-mp-layout
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to audit the generated HTML files for compliance and security.
- Evidence: SKILL.md provides specific validation commands:
xmllint --html --noout 目标.htmlandrg -n '@import|<script|<style|class=|id=|:before|:after|https?://' 目标.html. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted Markdown data from users, creating a potential surface for indirect prompt injection.
- Ingestion points: The agent reads user-provided Markdown files or text input to perform the conversion (documented in SKILL.md workflow).
- Boundary markers: No explicit delimiters or boundary markers for the untrusted content are specified in the conversion logic.
- Capability inventory: The skill can write files to the local filesystem (
公众号HTML输出/) and execute CLI tools for validation. - Sanitization: The skill implements "9 static checks" specifically designed to remove dangerous elements such as
<script>,<style>,class,id, and external URLs from the output HTML.
Audit Metadata