skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cwicr-report-generator/Gen Agent Trust Hub
cwicr-report-generator
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted user data (CSV, Excel, JSON) and interpolates it directly into HTML, Excel, and JSON reports without sanitization.
- Ingestion points: Data entering through QuickReport.from_dataframe or CWICRReportGenerator methods in SKILL.md.
- Boundary markers: No markers or instructions are used to delimit user data from report templates.
- Capability inventory: Filesystem write access via save_html and generate_excel methods in SKILL.md.
- Sanitization: No escaping or validation is performed on user-controlled strings before they are embedded in HTML output, presenting an XSS surface.
Audit Metadata