The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external websites to perform SEO audits, which could be used to deliver malicious instructions to the agent.
Ingestion points: WebFetch is used to retrieve rendered_dom, raw_html, and response headers from external URLs (SKILL.md).
Boundary markers: The instructions do not define clear delimiters or "ignore embedded instructions" warnings for the fetched content.
Capability inventory: The skill has access to Bash, Read, Grep, Glob, and WebFetch (SKILL.md frontmatter).
Sanitization: No explicit sanitization or validation of the fetched HTML or header content is described before processing.
[COMMAND_EXECUTION]: The skill requests access to the Bash tool to perform its operations. While common for development tasks, this tool provides a powerful execution environment that could be exploited if the agent is manipulated by untrusted input.
[EXTERNAL_DOWNLOADS]: The skill uses the WebFetch tool to download and analyze content from arbitrary URLs as part of its auditing function.

seo-indexability