park-investment-report
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted external sources via web searches to populate its investment reports. This creates a surface for potential prompt injection from websites, though the risk is inherent to the skill's primary purpose. \n
- Ingestion points: External data collected via
web.searchduring the '信息采集' (Information Collection) phase inSKILL.md.\n - Boundary markers: None explicitly defined in the instructions to separate external data from system instructions.\n
- Capability inventory: The skill has capabilities to perform web searches and write HTML files to the
@session/reports/directory.\n - Sanitization: There are no instructions for sanitizing or escaping the content retrieved from web searches before its inclusion in the HTML report.
- [EXTERNAL_DOWNLOADS]: The skill documentation specifies the use of 'Chart.js CDN' for rendering charts within the generated HTML reports. This is a standard reference to a well-known public library service used for data visualization.
Audit Metadata