geo-monitor
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches content from user-provided URLs (up to 10 pages) to perform its scoring analysis.
- [COMMAND_EXECUTION]: It invokes multiple subagents located in the relative filesystem (../geo-audit/references/agents/) to analyze different scoring dimensions in parallel.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by ingesting untrusted data from both external web pages and local baseline reports without explicit instruction boundary markers or sanitization logic. Mandatory Evidence Chain: 1. Ingestion points: Web content from user-specified URLs and local GEO-AUDIT-*.md baseline files; 2. Boundary markers: Absent in parsing logic for both web and local content; 3. Capability inventory: File-write operations for report generation, file-read operations for baselines, and parallel subagent execution; 4. Sanitization: No content filtering or escaping is specified before data processing.
Audit Metadata