The Agent Skills Directory

[NO_CODE]: The skill is composed strictly of metadata and markdown instructions for the AI agent. It does not include or reference any external scripts, binaries, or library dependencies.
[DATA_EXFILTRATION]: No network operations, hardcoded credentials, or access to sensitive local file paths (such as SSH keys or environment variables) were detected.
[PROMPT_INJECTION]: The skill prompt includes instructions for designing educational systems. It interpolates user-provided data such as skill_domain and knowledge_components. While these fields are not wrapped in formal delimiters (like XML tags), the skill has no access to tools or administrative functions that could be exploited via indirect injection.
Ingestion points: SKILL.md (Prompt section)
Boundary markers: Absent
Capability inventory: Restricted to text-based architectural design only; no tool access.
Sanitization: Absent
[SAFE]: The content is aligned with academic research in cognitive science (ACT-R theory) and presents no behavioral risks. The instructions are transparent and do not attempt to override agent safety filters or persist across sessions.

cognitive-tutoring-architecture-designer