The Agent Skills Directory

[SAFE]: The skill's primary function is to manage and synchronize coding standards within a repository. The instructions follow standard logic for reading, processing, and writing documentation based on developer input.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external local files to update project standards, which presents a surface for indirect instructions.
Ingestion points: unslop/SKILL.md and CODING_STANDARDS.md (via SKILL.md).
Boundary markers: None explicitly defined in the instructions for the content being processed.
Capability inventory: File read access to repository files and write access to CODING_STANDARDS.md.
Sanitization: The skill includes logic to ask the user for clarification on uncertain corrections and requires the agent to check diffs for duplicates, providing a level of human-in-the-loop validation.

ai-update-standards