The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the wordpress-pp-cli and wordpress-pp-mcp binaries from official GitHub and NPM repositories under the @mvanhorn organization. These resources are vendor-owned and central to the skill's intended functionality.
[COMMAND_EXECUTION]: Shell access is required to run the wordpress-pp-cli binary. The skill provides clear instructions and specific flags (e.g., --agent) to ensure command execution is non-interactive and produces structured output suitable for agent processing.
[INDIRECT_PROMPT_INJECTION]: The skill interacts with untrusted content from external WordPress sites via the REST API.
Ingestion points: Untrusted data is retrieved from posts, comments, media, and schema endpoints in SKILL.md.
Boundary markers: The skill utilizes the --agent flag to enforce JSON formatting, and the --select flag is recommended to narrow the scope of ingested fields.
Capability inventory: The environment allows shell command execution (Read Bash) and the CLI supports data delivery via local files or webhooks.
Sanitization: Data is processed as structured JSON, reducing the risk of the agent misinterpreting site content as instructions.

pp-wordpress