gds-test-framework
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script (
resolve_customization.py) located in the project's_bmaddirectory during activation and completion phases. - [COMMAND_EXECUTION]: The workflow executes configurable "activation steps" defined in
customize.tomland project-level overrides, providing a mechanism for dynamic command execution. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting data from the project root into the agent's context as foundational facts without sanitization.
- Ingestion points: Loads
config.yaml,project-context.md(via glob), and multiple.tomlfiles from the project directory. - Boundary markers: Absent. No delimiters or "ignore instructions" warnings are applied to the ingested content.
- Capability inventory: Local command execution via
python3and file system write access for scaffolding. - Sanitization: Absent. The skill does not filter or validate the contents of the loaded project files before processing them.
Audit Metadata