onenote-deploy-integration
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data, creating a potential surface for instructions embedded in that data to influence agent behavior.
- Ingestion points: The skill reads external documentation (referenced in
references/official-docs.md) as well as project-specific deployment artifacts, dependency locks, and configuration schemas. - Boundary markers: The skill implements significant manual safeguards, requiring explicit approval from release, content, and security owners before performing production actions or modifying sensitive configurations.
- Capability inventory: The skill is configured with
WriteandEdittools, which could be used to modify repository content based on potentially malicious data ingested during the review process. - Sanitization: There is no specific instruction for the automated sanitization or validation of the content within the deployment artifacts or documentation being read.
Audit Metadata