ai-seo
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill assists users in creating 'agent-ready' content designed for extraction by other AI agents. This introduces a surface for indirect prompt injection, where malicious instructions could be embedded in the optimized content or specialized machine-readable files.
- Ingestion points: Agent-readable files such as /pricing.md, llms.txt, and content-negotiated Markdown representations described in SKILL.md and references/agent-readiness.md.
- Boundary markers: The provided templates do not include specific delimiters or warnings to prevent agents from following embedded instructions.
- Capability inventory: The skill itself performs no file-writing or network operations on this data; the risk resides in the consumption of this content by external agents.
- Sanitization: No sanitization steps are suggested for the generated machine-readable content.
- [EXTERNAL_DOWNLOADS]: The documentation suggests auditing sites using npx is-agentic. This command downloads and runs code from the npm registry.
- Evidence: 'npx is-agentic yourdomain.com' in references/agent-readiness.md.
- Context: The tool is provided by a well-known service (Vercel), making the reference informative rather than high-risk.
Audit Metadata