The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data by ingesting text and code files for quality analysis and automated cleanup. This creates a surface for indirect prompt injection if the processed content contains instructions that attempt to influence the agent's behavior during the review process.
Ingestion points: scripts/detect_slop.py and scripts/clean_slop.py read content from file paths provided as command-line arguments.
Boundary markers: The scripts do not implement specific delimiters or 'ignore embedded instructions' warnings when parsing input content.
Capability inventory: The skill possesses file-write and backup creation capabilities via scripts/clean_slop.py.
Sanitization: Content modification is performed using static regular expressions to identify and replace specific natural language patterns.
[COMMAND_EXECUTION]: The skill provides Python scripts that execute local file system operations, including reading files, writing modified content, and renaming files to create backups. These actions are restricted to the local environment and the specific files targeted by the user.

anti-slop