power-platform-architect
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an architect processing untrusted input from business requirements, use case descriptions, and meeting transcripts. This information is used to generate architecture recommendations and Mermaid.js diagrams which are then saved to the user's local directory.
- Ingestion points: User-provided transcripts and unstructured descriptions (found in SKILL.md under Phase 1).
- Boundary markers: No specific delimiters or instructions to ignore embedded commands within user data are provided.
- Capability inventory: The skill possesses the capability to write files to the local filesystem (Phase 5: 'save it to the user's computer as a .md file').
- Sanitization: The instructions do not define any sanitization or validation steps for external input before it is used to generate file content.
- [SAFE]: The skill references official documentation and well-known services which are documented neutrally.
- Evidence: References to 'https://learn.microsoft.com/en-us/connectors/connector-reference/connector-reference-powerautomate-connectors' (well-known service) and 'https://mermaid.ai/live/edit' (well-known visualization tool).
Audit Metadata