onenote-ci-integration
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill defines a defensive workflow for CI integration, focusing on documentation review and synthetic testing to avoid mutating real data or exposing secrets.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential ingestion surface as it requires the agent to process repository content including documentation, configuration, code, and test fixtures. * Ingestion points: Inspection of documentation, configuration, code, and fixtures (SKILL.md). * Boundary markers: The skill instructions explicitly call for the use of synthetic fixtures and redacted receipts to minimize risk. * Capability inventory: The skill uses standard file system tools (Read, Glob, Grep, Write, Edit) to analyze and update repository artifacts. * Sanitization: The workflow includes specific steps to ensure secrets and real notebook content are removed from fixtures.
- [EXTERNAL_DOWNLOADS]: The skill references official Microsoft Graph API documentation from the learn.microsoft.com domain. These are well-known, trusted resources used for informational purposes regarding API contracts.
Audit Metadata