company-tech-profile-zhcn
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, untrusted sources, which creates a surface for indirect prompt injection.
- Ingestion points: Technical data is retrieved from external websites, patent databases, and academic sources via tools such as Exa, Tavily, Brave, and Jina (referenced in SKILL.md and references/source-routing.md).
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the external content being processed.
- Capability inventory: The agent environment has capabilities for local file read/write (Markdown and CSV) and network search operations.
- Sanitization: There is no explicit requirement to sanitize or filter external content before it is interpolated into the final reports.
- [DYNAMIC_EXECUTION]: The skill utilizes external libraries to dynamically generate document deliverables from Markdown source.
- Evidence: In references/deliverables.md, the skill recommends using python-docx and reportlab for rendering findings into DOCX and PDF formats.
- [EXTERNAL_DOWNLOADS]: The skill framework relies on multiple external search and data retrieval services to function.
- Evidence: SKILL.md and references/source-routing.md detail the integration with Exa, Tavily, Brave, and Jina APIs for gathering technical evidence.
Audit Metadata