The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from several external and user-controlled sources, creating a surface for indirect prompt injection.
Ingestion points: Data is ingested from the WordPress REST API (WORDPRESS.md), server-maintained storefront integration guides (STORE.md), and local project configuration files like .hostinger/site.json (SKILL.md).
Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore potentially malicious instructions embedded within the ingested data.
Capability inventory: The skill possesses significant capabilities, including creating databases and writing server-side configuration files (DATABASE.md), provisioning websites and purchasing domains (SETUP.md), and deploying source code for building and execution (DEPLOYMENT.md).
Sanitization: The instructions do not specify sanitization or escaping requirements for data retrieved from external APIs before it is interpolated into agent prompts or configuration files.
[COMMAND_EXECUTION]: The skill instructs the agent to use curl commands to verify that deployed websites are live and that API endpoints (like the store checkout or WordPress REST API) are returning expected status codes and content.
[DYNAMIC_EXECUTION]: For Node.js applications, the skill manages a workflow where local source code is archived, uploaded, and then built and executed on Hostinger's servers, involving runtime compilation and process management.

hostinger-headless