seo-geo
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documentation includes an audit workflow that utilizes shell commands like
curlto fetch external content from potentially untrusted domains. This creates a vulnerability surface where an agent processing the command output could be influenced by malicious instructions embedded in those external files.\n - Ingestion points: SKILL.md (via shell commands for auditing robots.txt and sitemaps)\n
- Boundary markers: Absent; the skill does not instruct the agent to use delimiters or ignore embedded instructions within the fetched data\n
- Capability inventory: Shell execution capabilities for
curl,grep, andhead\n - Sanitization: Absent; the external content is processed as raw text\n- [COMMAND_EXECUTION]: The skill documentation provides specific shell commands for technical SEO auditing. While these are common diagnostic tools, they define a capability for the agent to interact with the shell and network to retrieve and inspect remote resources.
Audit Metadata