The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses several shell commands to interact with the file system, including ls, find, du, mkdir, and mv. These commands are executed based on user-provided target directories to inspect structures, generate hashes, and move files.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local file system which could influence the agent's behavior.
Ingestion points: The skill reads file names, types, and sizes through ls -la, find -exec file {}, and du within the SKILL.md instructions.
Boundary markers: There are no specific instructions or delimiters provided to the agent to treat filenames or file metadata as potentially untrusted data.
Capability inventory: The agent has the capability to create directories (mkdir), move files (mv), and delete files (via the cleanup and duplicate removal logic described in SKILL.md).
Sanitization: The skill lacks sanitization or escaping mechanisms for file paths and metadata before they are used in organizational logic or displayed to the user.

file-organizer