overseas-patent-translation
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted patent documents provided by the user, which presents a surface for indirect prompt injection attacks.
- Ingestion points: The skill ingests user-uploaded or entered Chinese patent materials (claims, descriptions, abstracts, etc.) as specified in the 'Input Processing Workflow' section of
SKILL.md. - Boundary markers: The instructions do not define clear delimiters (e.g., XML tags or triple backticks) or specific 'ignore embedded instructions' warnings when processing the source text.
- Capability inventory: The agent has the capability to write files to the local file system (using the project folder structure defined in
SKILL.md) and invoke multiple external Model Context Protocol (MCP) tools for searching and mining patent data as described inREADME.md. - Sanitization: There is no evidence of sanitization, escaping, or filtering of the input content before it is processed by the translation logic or potential tool calls.
Audit Metadata