The Agent Skills Directory

[SAFE]: The skill serves as a set of developer guidelines for code quality and architectural simplicity. The instructions and examples provided are standard best practices for maintaining clean codebases and do not contain any malicious instructions, bypass attempts, or unauthorized command execution patterns.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external codebases using Read and Grep tools. While it lacks explicit boundary markers for untrusted data, its primary function is structural analysis and reporting, with no patterns suggesting the automated execution of analyzed content.\n
Ingestion points: Local project files and codebase directories accessed via Read and Grep tools (SKILL.md).\n
Boundary markers: None explicitly defined for isolating code content from analysis instructions.\n
Capability inventory: The skill is permitted to use Bash, Read, and Grep for searching and inspecting files.\n
Sanitization: Not applicable as the skill does not demonstrate behavior involving the interpolation of analyzed data into high-risk prompts or executable shell commands.

minimal-abstractions