AI agents call get_track_midi to retrieve information from Orpheus without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and retrieves MIDI data from an audio project without creating, modifying, or deleting any information. It is a pure read operation with no destructive or executable side effects. The blast radius of misuse is minimal—an AI agent reading MIDI data poses no risk to data integrity or system security.
From the tool's definition Tool description explicitly states 'Read a take's MIDI notes' and retrieves metadata (pitch, start/duration, velocity) and CC data with no modification or side effects.
Attacks that exploit this kind of access
Read a take's MIDI notes (pitch, start/duration in BEATS, velocity) + CC. It is categorised as a Read tool in the Orpheus MCP Server, which means it retrieves data without modifying state.
Register the Orpheus MCP server in PolicyLayer and add a rule for get_track_midi: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Orpheus. Nothing to install.
get_track_midi is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_track_midi rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for get_track_midi. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
get_track_midi is provided by the Orpheus MCP server (mal0ware/orpheus). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →