The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute WP-CLI (wp) commands for WordPress operations such as database exports, search-replace, and plugin management. This is the core functionality intended for the skill.
[DYNAMIC_EXECUTION]: The scripts/wpcli_inspect.mjs script utilizes node:child_process.spawnSync to invoke the wp binary at runtime. This execution is restricted to the specific wp command and its inspection subcommands (--info, core version, etc.).
[PRIVILEGE_ESCALATION]: The skill documentation and the wpcli_inspect.mjs script provide support for the --allow-root flag, which allows WP-CLI to run with superuser permissions. The documentation correctly identifies this as a potential security risk and instructs users to avoid it unless necessary for the specific environment.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes data from external WordPress installations.
Ingestion points: The scripts/wpcli_inspect.mjs script reads WordPress options (e.g., siteurl, home) and core version information using WP-CLI.
Boundary markers: The script outputs data in structured JSON format but does not include explicit "ignore instructions" delimiters for the values retrieved from the database.
Capability inventory: The skill has the capability to execute system commands via spawnSync and perform database operations through WP-CLI.
Sanitization: The script trims whitespace from tool output but does not sanitize the content of the database options before returning them to the agent context.

wp-wpcli-and-ops