[COMMAND_EXECUTION]: The skill makes extensive use of over 50 shell and Python scripts to automate development workflows. These scripts perform various tasks such as git operations, filesystem management, and status reporting. This behavior is central to the tool's intended purpose as a developer toolkit.
[EXTERNAL_DOWNLOADS]: Several scripts, including mb-upgrade.sh and mb-version-check.sh, connect to well-known and trusted services like GitHub and PyPI to check for software updates. This is a standard and expected feature for maintaining an up-to-date installation.
[INDIRECT_PROMPT_INJECTION]: The skill's primary feature is ingesting project-specific markdown files (such as status, roadmaps, and plans) from the .memory-bank/ directory into the AI agent's context. While this creates an ingestion surface for potentially untrusted data from a repository, the skill is designed to manage this project state transparently.
[DYNAMIC_EXECUTION]: The toolkit includes a structured test runner (mb-test-run.sh) that dynamically executes project-level test suites. It also utilizes Python's ast module and standard library subprocess calls for internal utility scripts and codebase mapping.

memory-bank