The Agent Skills Directory

[COMMAND_EXECUTION]: Includes a shell script scripts/audit-middleware.sh used to analyze Express source code for middleware patterns and route definitions. The script is restricted to the project directory and performs read-only operations using standard tools like find and grep with safe handling for filenames containing spaces.
[PROMPT_INJECTION]: The instructions in SKILL.md define a structured three-phase process (THINK, PLAN, EXECUTE) and include explicit security rules to prevent common attack vectors. It specifically forbids eval, curl | bash, and silent dependency installations.
[SAFE]: The skill operates entirely within the user's project directory and relies on standard package management. It includes comprehensive documentation on execution order preservation and error handling to ensure a safe migration that maintains the existing API contract.
[SAFE]: The skill processes user code (Express source files) which is an ingestion point for potential indirect prompt injection. However, the skill mitigates this through a structured planning phase and a verification checklist that ensures the final output matches the original API contract.

express-to-nestjs-migrator