The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of analyzing untrusted external data.
Ingestion points: In SKILL.md, the agent is instructed to fetch external website content using web_fetch and curl. It also reads local project files such as .agents/product-marketing.md and processes external data exports like Screaming Frog files.
Boundary markers: The instructions lack specific boundary markers or 'ignore' directives to prevent the agent from following instructions potentially embedded in the audited website's content.
Capability inventory: The agent uses web_fetch and curl to access remote data and reads local filesystem context.
Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from external sources before it is processed by the LLM.

seo-audit