The Agent Skills Directory

[SAFE]: The skill provides best practices and principles for system architecture and design. It is composed entirely of natural language instructions and does not perform any programmatic actions.
[NO_CODE]: There are no scripts, executables, or command-line instructions within the skill files.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill is designed to process user-provided system requirements, constraints, and quality attributes (SKILL.md).
Boundary markers: Not present.
Capability inventory: None. The skill has no access to the file system, network, or shell execution.
Sanitization: Not present.
Assessment: While the skill processes untrusted data (user requirements), it lacks any executable capabilities that could be abused through prompt injection.

solution-architecture