The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script located at {project-root}/_bmad/scripts/resolve_customization.py during activation. This script is used to merge configuration data from customize.toml and other project-specific override files into the active workflow environment. While the script is local to the project, this represents shell command invocation based on file-system state.\n- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its core function of ingesting and analyzing external project documents.\n
Ingestion points: The workflow identifies and reads the full contents of PRD, Architecture, Epic, and UX documents from the {planning_artifacts} directory as described in steps/step-02-prd-analysis.md and steps/step-03-epic-coverage-validation.md.\n
Boundary markers: The instructions lack explicit delimiters or safety guardrails to prevent the agent from following instructions that might be embedded within the project documents being analyzed.\n
Capability inventory: The skill environment permits shell command execution (python3) and file system read/write operations within the project scope.\n
Sanitization: There is no evidence of content validation or sanitization applied to the extracted text before it is processed as requirements for the final assessment report.

bmad-check-implementation-readiness