competitive-intel-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 exhibits an indirect prompt injection surface due to the way it processes user input for report generation.
- Ingestion points: User-provided inputs (e.g.,
OUR_COMPANY,INDUSTRY) enter the agent context and are passed as environment variables toscripts/generate_report.py. - Boundary markers: Absent. The skill does not use specific delimiters or instructions to prevent the agent from interpreting embedded commands in input data.
- Capability inventory: The skill has file-writing capabilities via
scripts/generate_report.py. - Sanitization: Absent. The script performs simple string replacement to populate HTML templates (
references/template_v12.html). Because inputs are not HTML-escaped, malicious input containing script tags could be executed when the user opens the generated report. - [EXTERNAL_DOWNLOADS]: The skill references and downloads resources from established external sources.
references/template_v8.htmlloads theChart.jslibrary from thecdn.jsdelivr.netcontent delivery network.- Templates include hyperlinks to patent documentation on the vendor-owned
eureka.zhihuiya.comportal.
Audit Metadata