The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the uv run utility to execute local scripts (resolve_customization.py, resolve_config.py, and memlog.py) stored in the {project-root}/_bmad/scripts/ directory as part of its standard activation and logging workflow.
[DYNAMIC_EXECUTION]: The agent is instructed to dynamically execute steps defined in configuration variables (activation_steps_prepend, activation_steps_append) and load data from file paths specified in {workflow.persistent_facts}, creating a flexible execution environment.
[EXTERNAL_DOWNLOADS]: The skill references a remote repository for its core functionality and suggests that users install missing components via npx skills add bmad-code-org/BMAD-METHOD, which involves downloading code from a vendor source.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection by processing external source materials. Ingestion points: The Discovery section explicitly invites 'brain dumps' and processes external source material such as Slack threads, meeting transcripts, and research subagent digests. Boundary markers: The instructions lack explicit delimitation or warnings to ignore instructions embedded within the ingested source material. Capability inventory: The skill has the ability to execute shell commands (via uv run), write to the local filesystem (briefs and logs), and invoke MCP tools for external data transfer. Sanitization: No sanitization or filtering logic is provided for the aggregated external data.
[DATA_EXFILTRATION]: The skill is designed to route finalized artifacts to external systems like Confluence or Notion using MCP tools defined in the external_handoffs configuration, representing a managed data movement capability.

bmad-product-brief