The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from external websites using the WebFetch tool, which constitutes a potential attack surface for indirect prompt injection.
Ingestion points: The skill relies on the WebFetch tool to retrieve the HTML content of a URL provided by the user (SKILL.md).
Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore potential malicious prompts embedded within the fetched HTML.
Capability inventory: The skill is restricted to WebFetch and Read tools via the YAML frontmatter, limiting the potential impact of an injection to the current session's analysis output.
Sanitization: The skill does not specify any sanitization, filtering, or escaping of the external content before it is processed by the agent.
[SAFE]: The skill's behavior is entirely consistent with its stated purpose of SEO analysis. It does not attempt to access sensitive local files, establish persistence, or execute unauthorized commands. The metadata and instructions are transparent and follow best practices for the intended use case.

30x-seo-page