The Agent Skills Directory

[SAFE]: The skill is entirely instructional, providing guidance on Clean Architecture, Domain Driven Design, and efficient library usage. It does not include any scripts, commands, or network operations that could pose a security risk.- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it is intended to process user-provided requests for architectural advice (Ingestion points: SKILL.md example). It does not implement specific boundary markers or instructions to disregard potential commands within that input (Boundary markers: Absent). However, the skill lacks any functional capabilities such as file system access, network operations, or shell execution (Capability inventory: None), and it does not perform sanitization of inputs (Sanitization: Absent). Consequently, the lack of dangerous tools makes this surface non-exploitable.- [METADATA_POISONING]: The YAML frontmatter includes a 'risk: critical' label. This metadata is inconsistent with the skill's actual function as a safe instructional resource, though it does not represent a functional security vulnerability.

software-architecture