multi-lang-readme-cn
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests content from external files (READMEs) which could contain hidden instructions targeting the LLM.
- Ingestion points: Processes full content of README.md and other local files as specified in the '读取内容' (Read Content) section of SKILL.md.
- Boundary markers: There are no instructions to use delimiters or ignore potentially malicious instructions within the source documents.
- Capability inventory: Includes file reading and writing capabilities to the local environment.
- Sanitization: Lacks explicit sanitization, validation, or filtering of the content being translated and rewritten.
Audit Metadata