The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests and processes textbook chapter content from project directories (e.g., docs/chapters/). Malicious instructions embedded within these markdown files could potentially influence the agent's output or behavior during quiz generation. The skill does not implement explicit boundary markers or sanitization logic to mitigate this ingestion risk.
[SAFE]: The skill performs authorized file system operations consistent with its stated purpose, such as reading course descriptions and learning graphs and writing quiz markdown, JSON metadata, and quality reports. It uses standard system commands for timestamping and logging without accessing sensitive environment variables or private system paths.
[SAFE]: There is a documented inconsistency between the README and SKILL.md regarding parallel execution; the README (v0.3) describes it as a default feature for multiple chapters, while the SKILL.md (v0.4) explicitly forbids it to ensure token efficiency and prevent race conditions. The agent operates under the more restrictive serial processing model defined in its core instructions.

quiz-generator