io.github.AudialAI/audial-mcp
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AUDIAL_API_KEY | Yes | Your Audial API key, required. From https://audialmusic.ai. | |
| AUDIAL_USER_ID | Yes | Your Audial user id, required. From https://audialmusic.ai. | |
| AUDIAL_RESULTS_DIR | No | Directory where Audial results are downloaded. | ~/Audial |
| AUDIAL_API_BASE_URL | No | Audial API base URL. Defaults to the production API. | |
| AUDIAL_JOB_TIMEOUT_S | No | Job timeout in seconds. | 900 |
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| stem_splitC | Split a track into separate instrument stems, optionally retimed and rekeyed. |
| analyzeA | Analyze a track: BPM, key, loudness and other characteristics. Results are in |
| segmentB | Detect song sections (intro, verse, chorus...) and analyze components within them. |
| masterB | Master a mix, optionally matching a reference track. |
| generate_samplesB | Extract a sample pack (one-shots and loops) from a track. |
| generate_midiC | Transcribe an audio file to MIDI. |
| generate_musicC | Generate music with the Audial music model. Text to music, covers, remixes, stem extraction, completion, or analysis (understand). |
| sound2vitalA | Turn a one-shot sample into an editable Audial Synth preset. |
| text2voxB | Synthesize a sung vocal (and MIDI) from lyrics, a melody and a reference voice. |
| list_resultsA | List previous Audial results in the results folder, newest first. |
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 10 tools
Most tools cover distinct audio operations such as stem splitting, analysis, mastering, MIDI transcription, and vocal synthesis, so many choices are clear. However, generate_music advertises stem extraction, analysis, and completion, overlapping with stem_split, analyze, and segment, which could cause misselection.
Names follow mixed conventions: verb-only (analyze, segment, master), verb_noun (generate_samples, list_results), and concatenated forms (stem_split, sound2vital, text2vox). They remain readable, but there is no predictable naming pattern.
With 10 tools, the set sits in the recommended 3-15 range and each tool represents a distinct audio or music capability. The surface does not feel bloated or thin for this domain.
Core workflows are covered: splitting, analysis, segmentation, mastering, sample/MIDI generation, music/vocal generation, synth preset conversion, and listing results. A minor gap is the lack of get_result/status or deletion/management operations beyond list_results.