The Agent Skills Directory

[SAFE]: No security issues detected. The skill instructions focus on pedagogical effectiveness, grounding quizzes in primary sources, and maintaining user-transparent state through local markdown artifacts. It explicitly guides the agent to avoid memorization trivia and to respect the user's workflow by not dirtying the repository worktree.
[INDIRECT_PROMPT_INJECTION]: The skill inherently possesses an attack surface for indirect prompt injection because its core functionality requires reading and analyzing external codebase files and specifications. These sources could theoretically contain malicious instructions designed to influence the agent. However, the risk is mitigated as the skill's activities are confined to local file reading and generating non-executable documentation.
Ingestion points: Primary sources including codebase files, design documents, and specifications (Phase 1, Step 1).
Boundary markers: No specific boundary delimiters are defined for the source content being analyzed.
Capability inventory: File system access for reading source code and writing markdown progress trackers to the session artifact directory or temporary storage.
Sanitization: No explicit sanitization or filtering of external file content is described in the instructions.

quiz-me