The Agent Skills Directory

[COMMAND_EXECUTION]: The activation protocol executes a local vendor script: node .aiox-core/development/scripts/generate-greeting.js qa.
[COMMAND_EXECUTION]: Starter commands such as *code-review and *review-build trigger automated processes that execute logic based on user-supplied parameters like {story} or {scope}.
[COMMAND_EXECUTION]: The command *review-build is configured to write a report file named qa_report.md to the local file system.
[PROMPT_INJECTION]: The skill processes untrusted user-supplied data in its commands, creating an attack surface for indirect prompt injection. Ingestion points: {story} and {scope} arguments; Boundary markers: absent; Capability inventory: command execution (node) and file-write operations; Sanitization: none identified in the current configuration.

aiox-qa