The Agent Skills Directory

[SAFE]: The skill is designed to assist with domain modeling by maintaining a glossary (CONTEXT.md) and recording architectural decisions (ADRs). It promotes best practices like lazy file creation and user confirmation before writing significant records.- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill only interacts with local project files relevant to its purpose, such as CONTEXT.md and ADR markdown files. There are no network requests, credential harvesting, or access to sensitive system directories.- [REMOTE_CODE_EXECUTION]: No external scripts are downloaded or executed. The skill contains no package installation commands or dynamic code execution patterns.- [INDIRECT_PROMPT_INJECTION]: The skill processes project content and user input to update documentation. While this is an ingestion surface, the risk is addressed by the skill's focus on structured documentation and the requirement for explicit user approval for ADR creation.

domain-modeling