bmad-architecture-design
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a local Python script (
scripts/generate_architecture.py) to automate the rendering of architectural documentation from structured JSON data. The script utilizes standard Python libraries and performs file operations exclusively within the workspace. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external documents, including
docs/PRD.mdanddocs/epics.md. While the skill lacks explicit boundary markers for these inputs, its functionality is limited to generating documentation, and it uses safe templating for output assembly. - Ingestion points: The agent is instructed to read project documentation (
docs/PRD.md,docs/epics.md), and the generation script loads structured data from a temporary JSON file. - Boundary markers: No explicit delimiters or warnings for embedded instructions are specified in the reading process.
- Capability inventory: The skill uses platform-provided tools for reading and writing files and executes a local script to generate markdown artifacts.
- Sanitization: The agent summarizes high-level requirements, and the Python script uses
string.Templatefor property substitution to avoid standard injection vectors.
Audit Metadata