The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external project artifacts such as setup-requirements.md to drive its provisioning logic. This ingestion of untrusted data creates a surface for indirect prompt injection where a compromised artifact could influence the agent to perform unauthorized site modifications or data operations.
Ingestion points: Reads and verifies the contents of migrations/<project>/setup-requirements.md.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore instructions' warnings when parsing the requirements file.
Capability inventory: The skill is capable of performing network modifications via CallWixSiteAPI and executing shell commands for tool configuration.
Sanitization: There are no explicit validation or sanitization steps defined for the content of the ingested markdown file.
[COMMAND_EXECUTION]: The skill provides procedures for installing and configuring ngrok using shell commands such as brew install ngrok and ngrok config add-authtoken. While provided as an optional step for media reachability verification, the use of tunneling tools involves executing external binaries and creating public network endpoints for local services.

rp-execute-setup