The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted web content (HTML and DOM) to perform SEO audits, creating a potential surface where malicious instructions embedded in external web pages could attempt to influence the agent's actions.
Ingestion points: The skill reads rendered_dom and raw_html snapshots during the audit process.
Boundary markers: The instructions lack explicit boundary markers or instructions to the agent to isolate the untrusted HTML from its primary command context.
Capability inventory: The skill utilizes the Bash tool for script execution and WebFetch for data retrieval.
Sanitization: No sanitization, filtering, or validation of the ingested HTML content is mentioned.
[COMMAND_EXECUTION]: The verification section provides a command for the agent to execute a local Node.js script (node ${CLAUDE_SKILL_DIR}/../../scripts/parse-html.mjs) using the Bash tool. While this is an internal script within the environment, it represents the execution of shell commands as part of the skill's workflow.

seo-mobile