The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest data from a local file if present. Ingestion points: .agents/product-marketing-context.md (as referenced in SKILL.md). Boundary markers: Absent. Capability inventory: No file-writing, network operations, or shell command execution detected across the skill instructions. Sanitization: Absent. The potential risk is mitigated by the skill's lack of actionable tools or system-level capabilities.
[SAFE]: The skill is a documentation-only asset focused on information architecture guidelines. It contains no executable scripts, external package dependencies, or network-bound functionality. It generates passive text outputs like ASCII trees and Mermaid diagrams.

site-architecture