architecture-diagrams
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFE
Full Analysis
- Infrastructure File Ingestion: The skill analyzes files such as Terraform, Bicep, and Kubernetes manifests to generate architectural visualizations. While these files are external inputs, the skill's capabilities are limited to generating ASCII and Mermaid text output without network or shell execution, which minimizes potential risks from untrusted data.
- Ingestion points: Infrastructure source files (Terraform, Bicep, ARM, shell scripts, K8s manifests, Docker files).
- Boundary markers: Not explicitly defined in the parsing instructions.
- Capability inventory: Text generation (ASCII/Mermaid) only; no network, subprocess, or file-write tools are utilized.
- Sanitization: Standard generative boundaries apply; the skill follows a specific output contract for formatting.
- Local Preference Persistence: It manages user preferences (like diagram format) by reading and writing to a local state file at
.copilot-tracking/architecture-diagrams/state.json. This is a routine method for maintaining a consistent user experience and does not involve access to sensitive system directories or credentials.
Audit Metadata