The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill accepts target input from the user via the $ARGUMENTS variable and includes it in shell commands. * Ingestion points: The target argument is parsed from $ARGUMENTS and used in the 'add' and 'remove' operations in SKILL.md. * Boundary markers: No delimiters or safety instructions are provided to prevent the agent from executing instructions embedded in the target names. * Capability inventory: The skill executes bash commands including cat, jq, mkdir, echo, and mv. * Sanitization: The instructions use jq --arg for JSON safety but do not provide guidance on escaping shell metacharacters in the input string, creating a command injection surface.
[COMMAND_EXECUTION]: The skill utilizes shell commands to interact with the local file system and manage its configuration. * Evidence: Bash scripts in SKILL.md perform file creation, modification, and display using common utilities.

scope