The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uv run to execute a local Python script located at {project-root}/_bmad/scripts/resolve_customization.py. This script is part of the vendor's provided tooling for configuration resolution.
[DYNAMIC_EXECUTION]: The instructions direct the agent to execute arbitrary steps defined in the activation_steps_prepend, activation_steps_append, and on_complete keys within configuration files (customize.toml and user-level overrides). This allows for the execution of command strings not defined in the skill's code.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its ingestion of multiple project-level files. 1. Ingestion points: Data is loaded from customize.toml, _bmad/custom/*.toml overrides, design-methods.csv, and config.yaml. 2. Boundary markers: Absent; there are no delimiters or explicit instructions to ignore potentially malicious embedded content within the ingested data. 3. Capability inventory: The agent has the capability to execute shell commands and write to the file system as instructed by the dynamic configuration. 4. Sanitization: Absent; no sanitization, escaping, or validation of the content is performed before it is used to direct the agent's behavior.

bmad-cis-design-thinking