The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user specifications during the design phase to determine project architecture and scaffolding commands.
Ingestion points: User-provided project specifications and technical requirements ingested during the DESIGN phase in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore malicious content within the user-provided data.
Capability inventory: The skill utilizes shell-based project scaffolding tools (dotnet, ng, ionic) and can trigger downstream build skills (domain-build) across files.
Sanitization: There is no mention of sanitizing or validating user inputs before they influence command parameters or architecture choices.
[COMMAND_EXECUTION]: The skill is designed to execute system-level commands to initialize new projects. This capability is driven by the interpretation of user requirements.
Evidence: Step 2 in SKILL.md explicitly instructs the agent to run commands such as dotnet new, ng new, and ionic start once the architecture is approved.

project-scaffold