The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified. The skill consists of markdown-based instructions that provide a template for architectural analysis and do not include any scripts or binary components.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted content from project documentation and source code (e.g., in docs/architecture/** and src/**). This represents an inherent vulnerability surface for indirect prompt injection if the processed files contain malicious instructions. However, the skill provides a structured framework for analysis and uses a restricted toolset (Read, Write, Grep, Glob), which limits the impact. 1. Ingestion points: docs/architecture/, docs/decisions/, src/**. 2. Boundary markers: Absent. 3. Capability inventory: Read, Write, Grep, Glob. 4. Sanitization: Absent.

well-architected