geo-report-pdf
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates an exposure surface where untrusted data from a local report can manipulate shell command arguments.
- Ingestion points: Extracts parameters such as brand_name, domain, and locations from GEO-AUDIT-REPORT.md.
- Boundary markers: Absent; there are no validation delimiters or instructions to prevent command interpretation.
- Capability inventory: Uses the Bash tool to execute pandoc and headless Google Chrome with dynamic parameters.
- Sanitization: Absent; the skill does not require escaping or sanitizing the extracted text before passing it to the command shell.
Audit Metadata