The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local script python3 {project-root}/_bmad/scripts/resolve_customization.py during activation and completion steps. This script is part of the vendor's project structure and is used to resolve and merge configuration settings.
[DYNAMIC_EXECUTION]: The skill supports an on_complete hook in customize.toml which, if populated by team or user overrides, is followed as a terminal instruction at the end of the workflow. This represents an execution path for dynamic instructions defined in configuration files.
[INDIRECT_PROMPT_INJECTION]: The skill has a notable attack surface for indirect prompt injection through its document analysis workflow. Ingestion points: The skill systematically searches for and reads GDD, Architecture, Epics, and UX documents in step-01 through step-05, as well as persistent_facts files defined in the configuration. Boundary markers: The instructions do not define clear delimiters or use 'ignore embedded instructions' warnings for the content read from these external documents. Capability inventory: The agent is authorized to execute shell commands (via the customization script) and write output to assessment reports. Sanitization: There is no evidence of sanitization, filtering, or validation of the content extracted from project documents before it is brought into the agent's context.

gds-check-implementation-readiness