The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs external diagnostic checks by executing shell commands such as curl and dig against user-specified domains.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted data from external sources and has the capability to modify website configurations.
Ingestion points: Reads website file content via hosting_files_website-content and runtime logs via hosting_nodejs_runtime-logs and hosting_nodejs_build-logs.
Boundary markers: The instructions do not define boundary markers or delimiters for the external content processed by the agent.
Capability inventory: The skill possesses significant modification capabilities, including deactivating WordPress plugins (wordpress_plugins_deactivate), changing PHP versions (hosting_php_update-version), and modifying Node.js build settings (hosting_nodejs_update-build-settings).
Sanitization: There are no instructions for the agent to sanitize or validate the content of the logs or files before using them to determine the cause of an issue or proposing a fix.

troubleshoot-website