The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function is to generate a local entrypoint script (script/build_and_run.sh) that automates multiple system operations including building code with swift or xcodebuild, terminating processes with pkill, and launching binaries via /usr/bin/open or direct invocation.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by ingesting untrusted data from the local project environment (such as scheme names, target names, and file paths) and interpolating them into executable shell scripts.
Ingestion points: The skill uses xcodebuild -list, find, and git rev-parse to discover project metadata.
Boundary markers: There are no explicit delimiters or warnings to ignore instructions embedded in the project files or metadata it processes.
Capability inventory: The skill has the capability to write files (cat >), modify permissions (chmod +x), and execute arbitrary shell commands via the generated build script.
Sanitization: The skill does not specify any sanitization or validation logic for strings extracted from project files before they are written into the build_and_run.sh script or used in shell commands.

build-run-debug