multi-lang-readme-cn
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a set of instructions for the agent to perform documentation translation tasks. It involves reading a source README file and writing translated content to new files within the project directory. No suspicious behaviors, obfuscation, or remote access patterns were found.
- [DATA_EXPOSURE]: The skill reads local project files to perform its primary function. This access is restricted to README documents and is used solely for the intended purpose of translation and documentation management.
- [PROMPT_INJECTION]: The skill includes an attack surface for indirect prompt injection as it processes content from potentially untrusted project documentation.
- Ingestion points: The skill reads root-level
README.mdfiles or other user-provided documentation. - Boundary markers: There are no specific delimiters or 'ignore' instructions defined to prevent the agent from obeying commands embedded in the source README.
- Capability inventory: The skill's capabilities are limited to reading and writing Markdown files.
- Sanitization: No explicit content sanitization is described in the workflow.
Audit Metadata