network-architecture-diagrammer
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection as it ingests untrusted external data from the local repository to build infrastructure diagrams.
- Ingestion points: The skill reads topology data from files such as AGENTS.md, docker-compose files, and repository documentation as specified in the SKILL.md workflow.
- Boundary markers: There are no explicit boundary markers or delimiters defined for the ingested data; however, the agent is instructed to normalize nodes into a specific set of categories.
- Capability inventory: The skill is capable of writing files to the local directory (e.g., .mmd files in docs/diagrams/) and executing a local Python validation script.
- Sanitization: The provided scripts/validate_mermaid_excalidraw.py script serves as a critical sanitization layer by checking the generated Mermaid output against a list of disallowed patterns, including click handlers, CSS classes, and HTML labels, which prevents the injection of interactive or malicious content into the final artifact.
Audit Metadata