The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to create local directories for reports using the mkdir -p command. While this involves shell execution based on user input, the skill includes a safety instruction to convert arguments into a "URL-safe folder name" to mitigate command injection risks.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of analyzing external website data.
Ingestion points: The skill ingests untrusted data from user arguments ($ARGUMENTS) and external website source code (meta tags, robots.txt, sitemaps) for audit purposes, as defined in the /audit-seo and /plan-seo command sections of SKILL.md.
Boundary markers: There are no explicit boundary markers or delimiters defined to separate untrusted website content from the agent's instructions during the reporting phase.
Capability inventory: The skill utilizes file system capabilities, specifically creating directories (mkdir) and writing markdown assessment reports to the local filesystem.
Sanitization: The skill includes a mitigation step instructing the agent to ensure target slugs are URL-safe before directory creation, though no specific sanitization is mentioned for the external content being placed into reports.

aholiab-seo