The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches and analyzes external content from URLs provided by the user via scripts/check_head.mjs and scripts/check_llms_txt.mjs. Maliciously crafted web pages could contain instructions designed to manipulate the agent's auditing process.
Ingestion points: Network fetches in the included Node.js scripts target arbitrary external URLs.
Boundary markers: The scripts do not implement specific delimiters or warnings to isolate external content from the agent's primary instructions.
Capability inventory: The skill allows network access and execution of local analysis scripts.
Sanitization: Content is parsed for SEO metadata but not sanitized for potential embedded instructions.
[COMMAND_EXECUTION]: The skill provides Node.js scripts (scripts/check_head.mjs and scripts/check_llms_txt.mjs) that are intended to be executed by the agent to perform technical SEO audits by fetching and parsing web content.
[EXTERNAL_DOWNLOADS]: The documentation suggests using npx lighthouse, which involves downloading and executing the Lighthouse auditing package from the public npm registry. This is an industry-standard tool provided by a well-known service.

seo-audit