The Agent Skills Directory

[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface due to its core functionality of processing external source code.
Ingestion points: The agent deeply inspects and processes untrusted source code, database artifacts, and metadata from the .import/ directory.
Boundary markers: The skill instructions lack specific delimiters or warnings to ignore potentially malicious embedded instructions within the imported source materials.
Capability inventory: The agent is authorized to perform file-system writes, execute shell commands via npm and node, and run automated tests using Playwright.
Sanitization: There are no documented steps for sanitizing or validating the imported content before it is used to derive the target application structure.
[EXTERNAL_DOWNLOADS]: The skill utilizes the npm package manager to download and install standard libraries for the required target stack (e.g., Vite, React Router, Drizzle).
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for project initialization, building, and automated verification using tools such as npm, node, and Playwright.

existing-project-migration