The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a locally bundled Python script (scripts/seo-scanner.py) to perform pattern matching across the project files. The execution is limited to read-only operations using the standard library modules os, pathlib, and re.
[DATA_EXPOSURE]: The tool processes codebase content but implements robust exclusion rules in collect_files to skip sensitive directories such as .git, node_modules, and environment folders (.venv), preventing accidental exposure of sensitive metadata or credentials.
[PROMPT_INJECTION]: While the skill ingests untrusted data (the user's source code), it lacks any mechanism to evaluate or execute that data. The scanner treats file content strictly as text for regex matching, mitigating the risk of indirect prompt injection.
[SAFE]: No external network requests, data exfiltration patterns, or obfuscation techniques were detected. The skill's behavior matches its stated purpose of SEO validation.

seo-validate