The Agent Skills Directory

[NO_CODE]: The skill is composed entirely of Markdown documentation, procedural guidelines, and templates. There are no executable scripts, binary files, or configuration files that execute code at runtime.
[SAFE]: The protocol implements human-in-the-loop governance. The AI is explicitly instructed to never auto-modify files and must instead present 'Proposed Domain Updates' for the user to review and apply manually, preventing silent data corruption or manipulation.
[SAFE]: The installation command npx skills add customware-ai/skills references a package repository associated with the skill's author ('customware-ai'), representing a standard vendor-provided installation method.
[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill's primary function is to ingest and trust content from project files (DOMAIN.md), which could be used to influence AI behavior if the files are modified by an external actor.
Ingestion points: Reads DOMAIN.md from the project root and supplementary markdown files within the docs/domain/ directory.
Boundary markers: Absent. The instructions do not define specific delimiters or 'ignore' commands for the content within the ingested domain files.
Capability inventory: The skill allows the agent to read file contents and propose structured text for file updates.
Sanitization: Absent. There is no specified validation or filtering for the business terminology or rules imported from the domain files.

domain-context