The Agent Skills Directory

[SAFE]: The skill instructions and templates contain no malicious patterns or dangerous instructions. All operations are focused on documentation generation and architectural analysis.
[COMMAND_EXECUTION]: The skill uses Glob and Grep tools to identify relevant code components within the local project. These tools are used for their intended purpose of codebase exploration and do not involve executing arbitrary shell commands.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading untrusted content from the codebase during its context gathering phase. Ingestion points: Project source code is read via Glob, Grep, and Read tools. Boundary markers: No specific delimiters or instruction-ignore markers are specified for the read content. Capability inventory: The skill has Read and Write access to the project file system to facilitate documentation generation. Sanitization: No explicit sanitization of ingested code content is mentioned before it is processed by the agent.

tech-design-doc