// SCAN REPORT

Midi

10 tools. 8 can modify or destroy data without limits.

8 write tools that can modify data. Rate limits recommended.

Last updated: 3 Jul 2026

8 can modify or destroy data

2 read-only

10 tools total

Community server · catalogue entry verified 03/07/2026

How to control Midi ↓

TOOL MAP

What Midi exposes to your agents

Read (2) Write / Execute (8) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous Midi tools

Execute · Medium open_midi_port Opening a MIDI port establishes an external connection/interface to hardware devices (synths, drum machines). Execute · Medium send_cc Sending a MIDI Control Change message executes an external operation that affects the state of connected hardware (synths, drum machines). Execute · Low send_note_off This tool sends a MIDI Note Off message to external hardware (synths, drum machines) via MIDI interfaces. Execute · Low send_note_on This tool sends a MIDI Note On message to external hardware instruments, causing them to produce sound.

8 of Midi's 10 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Midi

PolicyLayer is an MCP gateway — it sits between your AI agents and Midi, and nothing reaches the server without passing your rules. These are the rules we recommend:

Rate limit write operations

{
  "close_midi_port": {
    "limits": [
      {
        "counter": "close_midi_port_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "get_midi_status": {
    "limits": [
      {
        "counter": "get_midi_status_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Midi — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON MIDI →

Instant setup, no code required.

FULL CATALOGUE

All 10 Midi tools

EXECUTE 7 tools

Execute open_midi_port Open a MIDI output port by name. Use list_midi_ports to see available ports. Execute send_cc Send a MIDI Control Change message. Execute send_note_off Send a MIDI Note Off message. Execute send_note_on Send a MIDI Note On message. Execute send_pitchwheel Send a MIDI Pitch Bend message. Execute send_program_change Send a MIDI Program Change message. Execute send_sysex Send a MIDI System Exclusive message.

WRITE 1 tools

Write close_midi_port Close the currently open MIDI port.

READ 2 tools

Read get_midi_status Return current MIDI connection status. Read list_midi_ports List available MIDI output ports.

FAQ

Questions about Midi

How do I prevent bulk modifications through Midi? +

The Midi server has 1 write tools including close_midi_port. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Midi.

How many tools does the Midi MCP server expose? +

10 tools across 2 categories: Read, Write. 2 are read-only. 8 can modify, create, or delete data.

How do I enforce a policy on Midi? +

Register the Midi MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Midi tool call.

Deterministic rules across all 10 Midi tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MIDI →

Instant setup, no code required.

10 Midi tools catalogued and risk-classified — across an index of 43,000+ MCP servers.