The Agent Skills Directory

[COMMAND_EXECUTION]: Executes local Python scripts (resolve_customization.py and resolve_config.py) located within the project's _bmad/scripts/ directory to manage configuration and customization layers.
[COMMAND_EXECUTION]: Designed to execute arbitrary shell commands defined in the activation_steps_prepend and activation_steps_append configuration blocks, which can be modified via team or user-level configuration files.
[PROMPT_INJECTION]: Susceptible to indirect prompt injection (Category 8) as it processes various project documents that could contain untrusted data.
Ingestion points: Reads epics, story files, architecture documentation, and product requirements (PRDs) from the project directory.
Boundary markers: Missing explicit delimiters or instructions to ignore embedded commands when processing these files.
Capability inventory: Includes file-writing (saving retrospectives) and command execution (via helper scripts and activation steps).
Sanitization: No evidence of sanitization or validation of the content extracted from project files before it is used to influence agent behavior.

bmad-retrospective