gds-agent-game-dev
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The activation sequence in
SKILL.mdexecutes a Python script located at{project-root}/_bmad/scripts/resolve_customization.py. This script is used to merge configuration overrides and is a standard utility within the bmad-code-org framework. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it is designed to ingest and process project-specific documentation and configuration files.
- Ingestion points: The skill loads project context files via globs (e.g.,
project-context.md) and a globalconfig.yamlduring its activation steps. - Boundary markers: There are no specific instructions or delimiters mentioned in the loading process to isolate these external files from the agent's core instructions.
- Capability inventory: The agent is empowered to generate implementation code, author automated tests, and update sprint status files.
- Sanitization: The instructions do not specify any sanitization or validation routines for the external data being loaded.
Audit Metadata