The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple shell commands, including git status, a Python version check, and a release script. Specifically, the command python scripts/release.py $ARGUMENTS directly interpolates user input into the shell environment. This pattern is vulnerable to command injection if the input contains shell metacharacters like ;, &, |, or backticks.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data through the $ARGUMENTS variable.
Ingestion points: The $ARGUMENTS variable in SKILL.md captures user input.
Boundary markers: No delimiters or "ignore embedded instructions" warnings are used around the interpolated input.
Capability inventory: The skill has the capability to execute shell commands (git, python).
Sanitization: There is no evidence of validation, escaping, or filtering of the user-provided input before it is passed to the execution layer.

cut-release