The Agent Skills Directory

[SAFE]: The skill provides editorial guidelines for AI agents to process repository documentation and code comments. The instructions focus on brevity, clarity, and contract preservation without introducing malicious behavior.- [COMMAND_EXECUTION]: The workflow involves running git diff --check as a final verification step. This is a standard, safe development tool used to detect whitespace errors and unresolved merge conflicts in a repository.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external text data from the repository, creating an indirect prompt injection surface. However, the risk is negligible as the skill focuses on structural editing, and the impact is confined to the local text processing task.
Ingestion points: Markdown files, skill frontmatter, code comments, and strings within the target repository scope.
Boundary markers: None explicitly defined.
Capability inventory: File writing (authorized trim/edit tasks) and git diff execution.
Sanitization: None.

prose-standard