The Agent Skills Directory

[SAFE]: The skill provides instructional guidance for performance optimization without including executable scripts or automated network operations.\n- [SAFE]: All external references target official Microsoft Graph documentation on established domains (learn.microsoft.com).\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external evidence and documentation which represents a potential injection surface. 1. Ingestion points: Measured evidence files and Microsoft Graph documentation (references/official-docs.md). 2. Boundary markers: The skill explicitly requires the use of synthetic or approved non-production fixtures and redacted request evidence. 3. Capability inventory: Uses Read, Glob, Grep, Write, and Edit tools for repository artifacts. 4. Sanitization: Relies on human-in-the-loop auditable decisions and completeness proofs before applying changes.\n- [SAFE]: The skill incorporates explicit safety instructions, including the requirement for synthetic fixtures, redacted evidence, and approval boundaries before any testing.

onenote-performance-tuning