The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute Python scripts located within the project's local directory using the uv run command (e.g., {project-root}/_bmad/scripts/resolve_customization.py). These executions occur during the activation and validation phases of the workflow.
[DYNAMIC_EXECUTION]: The skill framework is designed to execute arbitrary shell commands provided through configuration arrays such as activation_steps_prepend, activation_steps_append, and on_complete. While intended for project-specific customization, this mechanism allows for the execution of arbitrary code if the local .toml configuration files are modified by an untrusted source.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from several files in the project environment to provide sprint summaries and risk assessments.
Ingestion points: sprint-status.yaml (Step 2), project-context.md (Step 1), and config.yaml (Step 4).
Boundary markers: The instructions do not define clear delimiters or specify that the agent should ignore instructions embedded within these data files.
Capability inventory: The skill possesses the ability to execute shell commands via uv run and perform file write operations to update status values.
Sanitization: There is no evidence of sanitization or strict validation of the natural language content within the ingested project files before it is used to influence the agent's logic.

gds-sprint-status