The Agent Skills Directory

[SAFE]: The skill provides static architectural guidelines, design templates, and technology recommendations. Analysis of SKILL.md, patterns.json, and structures.json revealed no evidence of prompt injection, data exfiltration, or obfuscation.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied requirements to generate designs, creating a surface for indirect injection. Ingestion points: functional and non-functional requirements in SKILL.md; Boundary markers: absent; Capability inventory: architecture advisory and document generation in SKILL.md (no active code execution or network operations); Sanitization: absent.\n- [SAFE]: The skill does not perform any network operations or external downloads, relying solely on local configuration files and referencing well-known technology services and frameworks.

system-architect