skills/microsoft/hve-core/mcsb/Gen Agent Trust Hub

mcsb

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • Reference-Only Architecture: The skill consists entirely of Markdown files and lacks any executable scripts, binaries, or automated tools. This design choice ensures that the skill provides informational value without introducing local execution risks.
  • Structured Data Retrieval: The lookup-playbook.md document provides a template for retrieving real-time security data from Microsoft Learn via a subagent. While this involves ingesting external data, the process is well-defined and scoped to official documentation sources.
  • Ingestion Points: External data is fetched at runtime from Microsoft's official documentation domains.
  • Boundary Markers: The skill suggests a structured Markdown format for the subagent's output to ensure clarity.
  • Capability Inventory: The skill contains no code and does not request specific tool permissions, relying on standard platform subagent capabilities.
  • Sanitization: The skill relies on the underlying agent platform for safe handling of retrieved text content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 11:39 PM
Security Audit — agent-trust-hub — mcsb