midi
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HF_TOKEN | No | Hugging Face token for downloading models, required if not logged in via hf auth login | |
| MIDI_MCP_OUTPUT_DIR | No | Directory for MIDI output and temporary downloads | <cwd>/.midi-output |
| MIDI_MCP_ALLOWED_INPUT_DIRS | No | Allowed local root directories for reading input, separated by system path separator | MCP process working directory |
| MIDI_MCP_DOWNLOAD_MAX_BYTES | No | Maximum size in bytes for a single local or remote audio file (200 MB default) | 209715200 |
| MIDI_MCP_MUSCRIPTOR_COMMAND | No | MuScriptor executable name or absolute path | muscriptor |
| MIDI_MCP_PROCESS_TIMEOUT_MS | No | Timeout for a single MuScriptor process in milliseconds (1 hour default) | 3600000 |
| MIDI_MCP_DOWNLOAD_TIMEOUT_MS | No | Timeout for the entire download in milliseconds (5 minutes default) | 300000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| audio_to_midiA | Transcribe a local or public HTTPS audio file to a multi-instrument MIDI file with MuScriptor. |
| check_modelA | Check whether MuScriptor and the writable output directory are ready without loading model weights. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 2 tools
The two tools serve completely different purposes: one is an environment/readiness check, the other is the core transcription operation. There is no overlap or ambiguity between them.
Both names use lowercase with underscores, but 'check_model' follows a verb_noun pattern while 'audio_to_midi' is a source_to_target pattern. The naming is clear and readable, with only a minor structural inconsistency.
With only two tools, the server feels minimal, but the tools directly cover the essential workflow for a focused MIDI transcription service. This is thin but not unreasonable for such a narrow scope.
The tool set covers the core workflow: check readiness and perform the audio-to-MIDI conversion. Minor gaps exist, such as no model management or configuration options, but the primary task is fully supported.