The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown instructions without any associated scripts or executable code.- [COMMAND_EXECUTION]: The skill requests file:read and file:write permissions. These are legitimate permissions for a tool designed to read raw transcript files and write cleaned versions back to the filesystem.- [PROMPT_INJECTION]: The skill processes untrusted user data (dictated text), which creates a surface for indirect prompt injection. However, since the skill is confined to text processing and does not have network access or system-level execution capabilities, the risk is negligible and handled by standard model guardrails.

dictation-cleanup