The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a local Python utility scripts/lint_style.py to analyze text drafts. This script only reads local files provided as arguments and uses Python's standard library for regular expression matching, with no network or write capabilities.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process and rewrite untrusted user-provided drafts, which creates a surface for indirect prompt injection. * Ingestion points: The skill ingests user content from .md, .html, and .txt files. * Boundary markers: No specific delimiters or 'ignore' instructions are mandated when the agent reads the draft content. * Capability inventory: The skill possesses the ability to execute local scripts and perform complex text generation/rewriting. * Sanitization: The lint_style.py script performs basic preprocessing (stripping code blocks and frontmatter) for its own analysis but does not sanitize the content passed back to the agent for rewriting.

humanize-content