The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided project requirements to generate source code, dependency manifests, and CI/CD configurations. This represents an indirect injection surface where malicious user input could potentially influence the generated project files or automation scripts.
Ingestion points: User requests for specific project templates, language setups, and distribution infrastructure.
Boundary markers: Not explicitly defined in the instructions for separating user input from generated templates.
Capability inventory: File writing (source, manifests, CI), command execution (build/test tools), and potential network operations (if authorization is granted for remote repos).
Sanitization: No specific sanitization or escaping of user-provided strings is mentioned.
[DYNAMIC_EXECUTION]: The skill is instructed to execute the generated setup and required check recipes (e.g., running cargo build, go test, or ruff check). While these are standard development tasks, it involves running commands on the host system based on the generated project structure.

project-bootstrap