The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a Node.js script located at scripts/wpcli_inspect.mjs which executes the wp command-line utility using spawnSync to inspect WordPress installation details such as version, site URL, and multisite status.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from a WordPress database (e.g., retrieving siteurl and home options), which could be leveraged to deliver malicious prompts if the target site is compromised.
Ingestion points: The wpcli_inspect.mjs script retrieves and outputs values from the WordPress database via WP-CLI option get commands.
Boundary markers: The inspection script returns values in a structured JSON object but lacks specific boundary markers or instructions for the agent to treat this data as untrusted content.
Capability inventory: The agent is empowered to perform sensitive and destructive actions including database imports, search-and-replace operations, and user/plugin management.
Sanitization: The script does not perform sanitization or validation on the data returned from the WordPress options table before displaying it to the agent.
[PRIVILEGE_ESCALATION]: Both the documentation and the wpcli_inspect.mjs script support the --allow-root flag. This flag instructs WP-CLI to ignore its internal safety check that prevents running as the root user, potentially exposing the host system to higher risk if the environment is misconfigured.

wp-wpcli-and-ops