The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external files like Game Design Documents (GDD) and incorporates user-provided narrative content into a generated design document. This generated content is later read back into the agent's context during workflow continuation or review steps.\n
Ingestion points: step-01-init.md (GDD files), step-01b-continue.md (existing narrative documents), and user responses across all story-building steps.\n
Boundary markers: The skill does not employ explicit delimiters or instructions to ignore potential commands embedded within the story content or GDD data.\n
Capability inventory: The skill utilizes shell command execution via uv run and extensive file system read/write operations across its lifecycle.\n
Sanitization: No sanitization or validation of the ingested narrative content is performed before interpolation into prompts or documents.\n- [COMMAND_EXECUTION]: The skill executes the local script {project-root}/_bmad/scripts/resolve_customization.py using uv run during the activation phase in SKILL.md and upon completion in step-11-complete.md. This script is part of the vendor's internal tooling for resolving configuration overrides and is executed with parameters derived from the local environment.\n- [DYNAMIC_EXECUTION]: The workflow includes an on_complete hook within customize.toml designed to execute arbitrary shell commands at the end of the narrative design process. The instructions in step-11-complete.md specifically direct the agent to execute whatever command is resolved from this configuration key.

gds-create-narrative