The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uv run to execute local Python scripts (resolve_customization.py and resolve_config.py) found in the {project-root}/_bmad/scripts/ directory. These scripts are invoked during the activation and completion phases to handle configuration logic.
[INDIRECT_PROMPT_INJECTION]: The workflow involves analysis of untrusted external documents to extract requirements, which could contain malicious instructions.
Ingestion points: Requirements extraction occurs in steps/step-01-validate-prerequisites.md from multiple project files, and SKILL.md loads persistent facts from files and globs.
Boundary markers: No specific boundary markers or 'ignore' instructions for the extracted content are mentioned in the process.
Capability inventory: The agent has the capability to execute shell commands via uv run, write to the filesystem, and invoke other skills like bmad-advanced-elicitation and bmad-party-mode.
Sanitization: The skill does not explicitly describe sanitization or filtering of the content extracted from the input documents.
[DYNAMIC_EXECUTION]: The skill contains logic in steps/step-04-final-validation.md to resolve a workflow.on_complete key from configuration and follow its value as a terminal instruction for the agent.

bmad-create-epics-and-stories