The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill's functionality is limited to processing text inputs and generating markdown outlines based on provided templates.
[SAFE]: An indirect prompt injection surface exists as the skill processes external context such as the Architecture Document, Research Dossier, and Author notes. However, this poses no functional security risk because the skill lacks network access, file system write permissions, or any code execution tools. * Ingestion points: SKILL.md (Session Start). * Boundary markers: Absent. * Capability inventory: None. * Sanitization: Absent.

chapter-architect