The Agent Skills Directory

[SAFE]: The skill's primary function is to facilitate architectural discussions and design exploration. It does not perform network requests, access sensitive system files, or execute arbitrary shell commands.\n- [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection as it ingests and analyzes user-supplied source code to generate design recommendations. 1. Ingestion points: User-provided codebase modules and candidates for deepening (DESIGN-IT-TWICE.md). 2. Boundary markers: Not specified; the agent processes the code as context for its sub-agents. 3. Capability inventory: Spawns sub-agents to generate code sketches and design trade-offs. 4. Sanitization: The skill does not implement specific sanitization for analyzed code snippets. This surface is inherent to the skill's purpose and handled through standard agent reasoning.

codebase-design