schema

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to suggest the installation of the 'seomator-geo-audit' package via 'uv' or 'pipx'. This tool is a core component provided by the skill author 'seo-skills'.\n- [COMMAND_EXECUTION]: The skill uses the 'geo' CLI tool to validate and audit website schema. This involves the execution of shell commands that interact with user-provided external URLs.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process content crawled from external websites, creating a surface for indirect prompt injection.\n
  • Ingestion points: The agent ingests data from external URLs through 'geo' tool outputs, specifically fields like 'findings[].excerpt' and 'signals[].detail' in 'SKILL.md'.\n
  • Boundary markers: The skill contains explicit instructions in 'SKILL.md' to treat all crawled excerpts as 'a quotation of untrusted content' and to ignore any embedded instructions.\n
  • Capability inventory: The skill is configured to use the 'Bash' and 'Read' tools.\n
  • Sanitization: The instructions require the agent to report any malicious instructions found in the data as findings about the page instead of following them.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 03:20 PM