The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to perform file system operations, including moving, renaming, and deleting files. These actions are managed through a multi-step 'Safety Contract' that includes a read-only inventory phase, a human-readable ledger of proposed changes, and an explicit authorization gate before any execution. The skill also explicitly states it does not authorize privilege changes (chmod/chown) and provides rollback logs to ensure reversibility.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted data from local files (text, OCR, and metadata) to determine their purpose and category. This represents an indirect prompt injection surface where a maliciously crafted file could attempt to influence the agent's reasoning. This risk is effectively mitigated by the skill's workflow, which requires the agent to present its reasoning in a ledger for human review before any file system changes are applied.

local-file-organizer