fumadocs-component-docs
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for generating documentation by interpolating component data into MDX templates, which creates a vector for malicious instructions in processed data.
- Ingestion points: Component metadata, such as
ComponentName, property descriptions, and usage examples, are ingested into the agent context via the templates defined inSKILL.md. - Capability inventory: The agent is tasked with generating and writing these values into the project's documentation files (e.g.,
content/docs/components/*.mdx). - Boundary markers: The skill does not provide clear delimiters or "ignore embedded instructions" warnings for the interpolated component content.
- Sanitization: There is no evidence of sanitization, escaping, or validation logic to process external content before it is used to generate documentation.
- [EXTERNAL_DOWNLOADS]: The skill includes documentation templates that generate commands for fetching packages from the
@8bitcnscope on the NPM registry. - [COMMAND_EXECUTION]: The skill provides templates for shell commands using
pnpm dlx(specificallypnpm dlx shadcn@latest add ...) intended for installation and project setup.
Audit Metadata