The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill provides an evaluate_python tool that allows the execution of arbitrary Python code within Ableton Live's internal environment. This is a powerful administrative capability that allows direct manipulation of the host application and could be misused if the agent is directed by malicious instructions.
[EXTERNAL_DOWNLOADS]: Installation requires cloning a repository from a third-party GitHub account (bschoepke/ableton-live-mcp) and installing dependencies via pip. This source is not verified as an official vendor or trusted organization.
[COMMAND_EXECUTION]: The documentation provides manual setup steps involving the execution of shell commands such as git clone and pip install -r requirements.txt, which download and install code from external sources.
[INDIRECT_PROMPT_INJECTION]: The skill has a potential surface for indirect prompt injection because it ingests data from the Ableton Live environment. 1. Ingestion points: The agent reads track information, device names, and captured audio data using tools like get_live_set_info and capture_audio. 2. Boundary markers: The documentation does not mention the use of delimiters or instructions to ignore embedded commands in the ingested data. 3. Capability inventory: The skill includes high-impact tools like evaluate_python and file insertion capabilities. 4. Sanitization: No mechanisms for sanitizing or validating data retrieved from the Ableton environment are described before the agent processes it.

ableton-live-mcp-control