ai-seo
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external content (blog posts provided as text or via URL), which creates a potential surface for indirect prompt injection attacks where instructions embedded in the analyzed content could influence the agent's behavior.
- Ingestion points: Untrusted blog content (text or URL) and the
.agents/blog-context.mdfile. - Boundary markers: The skill's instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when interpolating untrusted content into the prompt context.
- Capability inventory: The skill is restricted to reading local context files, performing web searches, and generating text-based editing recommendations. It does not have the capability to execute shell commands, perform unauthorized network operations, or modify sensitive system files.
- Sanitization: No evidence of input validation, filtering, or escaping for external content was found in the provided instructions.
Audit Metadata