The Agent Skills Directory

[SAFE]: The skill operates by analyzing the existing codebase and proposing refactoring opportunities based on architectural principles (depth, leverage, and seams). It uses documentation files (CONTEXT.md, LANGUAGE.md, ADR-FORMAT.md) to maintain consistent domain and architectural vocabulary.
[COMMAND_EXECUTION]: The skill utilizes an internal Agent tool with subagent_type=Explore to navigate the codebase. This is a standard procedure for AI agents to understand project structure and is used here in a read-only context for analysis.
[PROMPT_INJECTION]: No evidence of prompt injection or instructions to bypass safety filters was found. The instructions focus on architectural analysis and user-driven refactoring decisions.
[DATA_EXFILTRATION]: The skill does not perform any network operations or access sensitive credentials. It operates locally on the codebase files provided in the context.
[EXTERNAL_DOWNLOADS]: There are no external dependencies, package installations, or remote script fetches defined in the skill instructions.

improve-codebase-architecture