Moose Tracker
Server Details
Synth, sampler and music tracker for AI assistants: sound effects, UI sounds and songs as audio.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 9 tools
Most tools have clearly distinct roles: assets retrieves files, convert changes formats, render produces audio, sfx makes sound effects, lyrics analyzes text, and song plans/analyzes compositions. Adjacent overlap exists between song and render, and between describe and research_composition, but the detailed descriptions make the boundaries mostly clear.
The names are readable and mostly lowercase, but they mix nouns (song, lyrics, sfx, assets, feedback) with verbs (render, convert, describe) and one verb_noun name (research_composition). The convention is understandable but not predictable.
Nine tools is well within the appropriate range for a music and audio creation server. Each tool covers a distinct stage of the workflow: help, song planning, lyrics, sound effects, rendering, conversion, composition research, asset retrieval, and feedback.
The core creative lifecycle is covered: compose/plan/analyze songs, render audio, convert formats, generate sound effects, analyze lyrics, retrieve assets, get guidance, and send feedback. Minor gaps remain around explicit asset deletion/listing and project or song-library management, but agents can work around them.
Available Tools
9 toolsassetsRenew download linksARead-onlyIdempotentInspect
Use this when a download link was lost or is about to expire, or a render replied rendering: pass the asset ids from the result to get the links or the status, with no re-render and no charge. Files live about 2 h; verify:true re-checks a file's hash. An id that has expired says so: the same request again returns the same files (the synthesizer is deterministic). Schema a964948f.
Inputs: ids (required), verify.
convertConvert song formatsARead-onlyIdempotentInspect
Use this when a song should move between MTL text, a Moose Tracker .mtk project, a share link and standard files: MIDI (.mid) and MOD, XM, S3M or IT tracker modules, in either direction (files in as base64; a written file comes back as a download link). It takes a song into another program (a DAW, a notation app, another tracker) or brings one in; voices maps channels to sample packs or presets on import, and feel adds swing or humanize. Cues and macros are compiled away in .mtk and share links. Schema a964948f.
Inputs: to (required), from (required), input (required), feel, +2 more.
describeLearn Moose TrackerARead-onlyIdempotentInspect
Use this when the user asks for a sound effect, click, notification, jingle, audio, music, song or soundtrack, or mentions Moose Tracker: returns usage notes. No arguments gives a short index (workflow, limits, formats, next steps); topic opens one section: syntax, limits, styles, recipes, presets, mtl, instrument, effects, modulation, drums, loops, formats, sfx, examples, errors, genres, generators, samples, search, not_supported, workflow, volume, mix, sounds, character; q searches presets and the sample library; with topic="syntax", q opens one syntax section (q="arp") or a field. Song text up to 256 KB; one audio file holds about 8 min as WAV, longer as Ogg or Opus (exact seconds per format in the index). Tool list 1.6.1. Schema a964948f.
Inputs: q, genre, limit, topic, +3 more.
feedbackSend feedbackAInspect
Sends a note to the Moose Tracker team about a bug, a confusing tool, or a missing feature. Up to 20000 characters per call; longer reports can be split into parts. Notes are kept 1 year; include what happened, the call and its reply. Schema a964948f.
Inputs: type (required), message (required), mtl, ref, +2 more.
lyricsAnalyze lyrics for singingARead-onlyIdempotentInspect
Analyzes lyric text for setting to music: each word's syllables, stress and phones (ARPAbet), the rhyme scheme with perfect and near rhymes, scansion against a poetic meter, and where each syllable can start on a tracker rhythm grid for a time signature (stressed syllables on strong beats, a pickup for an unstressed start), with warnings for lines that do not scan. EXPERIMENTAL: the singing voice and lyric setting are early features; word clarity and fit vary from song to song, so render and listen before relying on a result. It writes no melody and no words of its own (fit mode only reshapes a melody you send): the assistant writes the notes, then puts each syllable on its note in MTL (C4'twin- D4'kle; '_ holds a syllable over more notes). Word setting: mode=plan returns, for every line, exactly how many notes it needs and their lengths as a melody template (x:2 x:2 x:4: write the tune on that rhythm); mode=fit with melody checks and fits the words on a melody you wrote, adapting it (adapt, default true): extra syllables split a note (split=repeat repeats the pitch, pass steps toward the next note), a line that starts unstressed on the downbeat gets a pickup note, extra notes carry the syllable (melisma '_), and the last note is held to its bar line (hold); every change is listed, and warnings name syllables that are crammed, stress-inverted or off the beat and phrases that are shorter or longer than the bars (bars). Each fit line returns a ready MTL cell string and the adapted melody. Pronunciation comes from the dictionary; words guessed by rule are listed (fix with say). melody: one line per lyric line (a "\n" between lines, or an array). Tokens split by spaces: a pitch (C5, F#4, Bb3; octave 0-9, C4 = 48), optionally with its length in rows (C5:4; default = step rows); "." a rest of one step, ".3" three steps, ".3r" three rows; "x" a placeholder pitch (from plan mode: x:2); "@-2" first token only: the line starts at row -2 (a pickup before the bar; default 0). A note lasts until the next token. Example: "C5 D5 E5:4 . G5:2". Up to 48 lines per call. Schema a964948f.
Inputs: text (required), bpm, rpb, say, +11 more.
renderRender a song to audioAIdempotentInspect
Produces WAV, Ogg or Opus files: the audio file a user asks for, to download or to embed in a game, app, presentation or video. A request for a song, a tune or music needs no file: song's share link plays it in the browser (the share link plays it, no render), and each render counts against the hourly render quota. Send mtl, ref (a songId) or compose; it checks the song itself (syntax: describe(topic="syntax"), sections by q). One file holds about 8 min as WAV (about 5 at 24-bit 48 kHz), longer as Ogg or Opus; a longer request is refused with the exact limit. len gives exact seconds; loop:true adds seamless loop points; stems:true gives one file per instrument role (@lead, @bass, @drums...); adaptive:true gives a zip of same-length looping layers in intensities, stingers from extra cues, a manifest, a Web Audio player and Godot, Unity, FMOD and Wwise notes; cues renders named sections. A long render replies rendering with asset ids, and assets(ids) gives the links when done. Files last 2 h; https://moosetracker.app/acceptable-use.html Schema a964948f.
Inputs: cue, len, mtl, ref, +19 more.
research_compositionResearch composition adviceARead-onlyIdempotentInspect
Finds composition advice. brief: once when starting a real song (sections, over ~20 s), not for simple sounds: advice with ranges for every stage (plan, harmony, melody, layers, sound, mix, review) from a curated corpus of about 50 styles; stage picks one, limit and avoid page. Then song(plan). q: before writing a melody, harmony, bass line, groove or arrangement, look up the music-theory guide (principle, MTL example, common mistakes; idioms for every genre id), e.g. q="melody feels static"; ids fetches entries song checks cite. Advice is retrieved: no AI composes music here. Schema a964948f.
Inputs: q, ids, key, role, +8 more.
sfxMake sound effectsAIdempotentInspect
Produces sound-effect files (up to 16 per call) to embed in an app, game, UI or video. Each item is one of: a role (press, release, toggle_on, toggle_off, continuous_tick, success, warning, notify, complete) with a shared family (seed, character, root, scale) for a coherent UI set with playback hints; a recipe (describe topic=recipes; recipes are starting points) tailored with pitch, bright, dur and seed, pairs alike; or an instrument line with play, which auditions one designed voice. Kinds mix in one call. revise patches a family by its handle. Reports design_mode and loudness; WAV by default, Ogg or Opus on request. Files last 2 h; https://moosetracker.app/acceptable-use.html Schema a964948f.
Inputs: zip, items, style, detail, +3 more.
songPlan, check and analyze a songARead-onlyIdempotentInspect
Plans and analyzes a song written in MTL text (render checks too); returns a share link that plays it in the browser (90 days, no render). Stages: plan, write, light analysis loop, one deep analysis, fix; then the share link to listen; one render only when the user asks for an audio file. song(plan) gives a skeleton; analyze:true: light (repeatable); deep:true: measured (once). Your own sound: i3 - name="Glass Lead" wave=saw cutoff=55 (a quoted word after iN is a preset); tracks play it as i3. The syntax: describe(topic="syntax") gives the core grammar and a list of sections, each read with q (q="synth" every synth field, q="arp", q="modulation" ...). Stopping a note: = releases it (its release plays out); an ECx effect cuts it dead after x ticks (_!EC0 on that step, D4!EC3 on the note). Each reply has a songId: send ref instead of the text again, with layers for batch edits (tracks, instruments, mix, order). It warns when a song is longer than one audio file holds (WAV about 8 min, Ogg or Opus longer). To work on a song in another program, convert it to MIDI or a tracker module. Schema a964948f.
Inputs: len, mtl, ref, alts, +16 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Browser DAW for AI assistants: build multitrack songs with MIDI, effects, stems and generation.
Audio for your agent: transcribe, speak, translate, summarise, plus sound effects and music.
Game sound effects from a text prompt: game-ready WAVs with seamless loops and variations.
Make AI Tracks, loops and sound effects, split stems, build drum kits and master songs.
Related MCP Servers
- AlicenseAqualityBmaintenanceEnables AI clients to compose original game music across 19 styles plus 32 sound effects and 11 looping ambiences, rendered deterministically to WAV (and optionally MIDI) from a seed and musical parameters. Everything is synthesized in pure Node.js with no dependencies, producing seamless loops, mono or multi-rate output, and reproducible renders for game soundtracks, menus, and environmental beds.4509 npmMIT
- AlicenseAqualityDmaintenanceEnables AI assistants to generate and control real-time audio synthesis through natural language descriptions using SuperCollider. Features 10 built-in synth types, pattern sequencing, audio recording, and server lifecycle management for creating sounds from simple English descriptions.112MIT
- AlicenseAqualityBmaintenanceGame sound effects for AI coding agents. 45 presets: Eastern instruments (guqin, pipa, erhu, temple bells, gongs, war drums), xianxia magic (sword qi, thunder, talismans) and general game sounds (UI, coins, combat, footsteps, ambience, monsters). Renders locally to WAV files or a dependency-free JavaScript snippet. No API key, no network, free for commercial use. Run with: npx -y sfxbrew mcp41,045 npmMIT
- FlicenseNot gradedqualityDmaintenanceEnables generating game assets from natural language prompts, including sprite sheets, animations, voiceover, sound effects, chiptune music, tilesets, backgrounds, VFX, and cohesive asset packs with shared style and audio DSP presets.-
Glama MCP Gateway
Add one secure layer between your agents and this server.