The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a parallel design exploration workflow in DESIGN-IT-TWICE.md that takes user-provided module constraints and dependency classifications to generate briefs for sub-agents. \n- Ingestion points: User-provided problem space descriptions, code sketches, and dependency categorization entering the agent context during the framing stage. \n- Boundary markers: The instructions do not define specific delimiters or instructions for sub-agents to ignore potential prompts embedded within the user-provided design constraints. \n- Capability inventory: The workflow utilizes the platform-native Agent tool to spawn sub-agents which may have access to further software development tools. \n- Sanitization: There is no specified validation or sanitization of the user input before it is interpolated into technical briefs for sub-agent execution. \n- [SAFE]: The core instructional content focuses on conceptual software design principles like deep modules and seams. The skill does not include executable code, perform network operations, or define external software dependencies.

codebase-design