gds-e2e-scaffold

Warn

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script at '{project-root}/_bmad/scripts/resolve_customization.py' to manage configuration merging from multiple project files.
  • [COMMAND_EXECUTION]: The workflow permits execution of arbitrary commands defined in the 'activation_steps_prepend', 'activation_steps_append', and 'on_complete' configuration fields. This allows for local command execution based on project-specific settings.
  • [PROMPT_INJECTION]: Potential for Indirect Prompt Injection via project context files. The skill ingests data from '{project-root}/**/project-context.md' as foundational context for the workflow.
  • Ingestion points: Reads project-specific context and configuration files like 'project-context.md' and 'customize.toml'.
  • Boundary markers: Absent; no specific delimiters are used to isolate ingested data from the system instructions.
  • Capability inventory: Includes shell command execution and broad file system write access in the project directory.
  • Sanitization: No validation or sanitization is performed on the ingested content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 7, 2026, 06:21 PM
Security Audit — agent-trust-hub — gds-e2e-scaffold