The Agent Skills Directory

[SAFE]: The skill files consist of instructional markdown content and templates for SEO auditing. Analysis across all threat categories indicates the content is benign and follows standard practices for agent-based tasks.
[PROMPT_INJECTION]: The skill defines a workflow for auditing external websites using web_fetch or curl, which represents an ingestion point for untrusted data (Indirect Prompt Injection surface). However, given the restricted scope of the audit task and the absence of high-risk capabilities like code execution or file-system modification, this surface does not pose a significant risk. No direct jailbreak or override instructions were found.
[DATA_EXFILTRATION]: The skill reads local context from .agents/product-marketing-context.md, which is a standard practice for agent personalization and does not involve accessing sensitive user credentials or system files.

seo-audit