geo-site-audit

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by fetching and processing untrusted content from external websites.\n
  • Ingestion points: scripts/geo_audit.py fetches data from URLs (robots.txt, llms.txt, and site HTML) provided at runtime.\n
  • Boundary markers: None; external content is formatted directly into the audit reports without delimiters.\n
  • Capability inventory: The skill performs network requests and writes output files to the local system.\n
  • Sanitization: No filtering or sanitization of instructions is performed on the retrieved data.\n- [DATA_EXFILTRATION]: The scripts/batch_audit.py script reads a local file path provided as an argument without validation, which could be exploited to read sensitive local system files via prompt injection.\n- [COMMAND_EXECUTION]: The skill's primary functionality is implemented through Python CLI scripts, requiring the execution of local commands.\n- [EXTERNAL_DOWNLOADS]: Integration documentation references external utilities like wkhtmltopdf and internal scripts like check_threshold.py that are not included in the skill distribution.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 02:33 AM