Riff ChatGPT Plugin
OfficialThis server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Riff ChatGPT Plugin
Turn your ideas into songs with vocals and instruments in ChatGPT. Write and revise lyrics, shape melodies and arrangements, listen to your music, and download the results. This repository packages Riff for ChatGPT.
Source and dependency
Extracted from Riff PR #32, source commit cd98f3f4191c2b1356d6048bfe44b3848d31cb2d. Its full Git ancestry is preserved so original authorship and the original integration remain inspectable. The extraction commit removes upstream implementation files from this repository's working tree.
Riff is a Git submodule at riff/, locked in upstream.lock.json to b7ab93cf22ffd9e95a7718aca2e0d451578411b1, the merge of standalone LoRA PR #28. This plugin owns MCP tools, the ChatGPT bridge and UI integration, shared drafts/settings, plugin metadata, packaging and host acceptance tests. Upstream Riff owns generation, LoRAs, recordings, the studio and native/control/media source pins.
git clone --recurse-submodules https://github.com/Flip-Engineering/Riff-ChatGPT-Plugin.git
cd Riff-ChatGPT-Plugin
npm ci
npx playwright install chromium
./scripts/validate.shValidation needs Python 3.10+, Node.js, Chromium and FFmpeg/FFprobe. Development validation runs on the homelab with TMPDIR and the checkout on NVMe. Python runtime packaging remains separately pinned for user distributions.
Related MCP server: AI Sound MCP Server
ChatGPT Chat target
ChatGPT Chat is the required user surface. The cloud distribution bundles the native engine for execution in the user's Chat environment. Private installation, native CPU generation, exact WAV export and inline player display have passed in Chat. Version 1.1.1 combines the native runtime, music skill and connected original Studio in one app-backed Riff plugin. Actual Chat checks cover library playback with the custom visualizer, seek and looping, shared score editing, and public LoRA installation and selection. The full studio transferred and played a complete 180-second WAV fixture; the separate standalone recording panel uses a labeled preview for larger files. See the measured results and limits. Writing and reviews use the current ChatGPT session. Studio writing, reviews and WAV export send requests directly to the current ChatGPT conversation from the original button event. The hosted bridge is an internal component of that plugin. The original local stdio distribution remains available.
Run and package
See plugin usage and runtime packaging. Build the UI with npm run build:ui. For development, start plugins/riff-chatgpt/server/main.py --riff-root "$PWD/riff" --workspace /absolute/private/workspace --ui-file "$PWD/plugins/riff-chatgpt/dist/studio.html".
Builders under plugins/riff-chatgpt/packaging/ consume the verified Riff submodule. scripts/package_chatgpt_plugin.py assembles the plugin-owned integration plus the curated pinned upstream runtime. Package receipts identify both the plugin commit and Riff commit. The complete studio and LoRA workflow are retained.
Update Riff by checking out a reviewed upstream revision inside riff/, updating upstream.lock.json, running validation and committing both pins together. Builders reject mismatched revisions and tracked changes. The UI adapter checks exact integration points and refuses incompatible upstream UI changes.
Delivery status
The cloud package is privately installed and tested; it has not been published to the public directory. Historical validation records pre-extraction evidence; downstream results are tracked in extraction validation and Chat acceptance. Platform package builds, real model inference, Apple Silicon behavior and actual ChatGPT host installation are reported independently. Music weights are downloaded separately under their own terms.
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP server for Producer/Riffusion AI music generation
Generate AI music via the Lacuna Music API from MCP clients like Claude Desktop & Code.
Direct AI video, image, music and storyboard production from any MCP client.
Generate images, video, audio and short films with 140+ AI models from any MCP client.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables interaction with Mureka's AI music generation APIs to create lyrics, songs, and background music through MCP clients like Claude Desktop and OpenAI Agents.MIT
- FlicenseNot gradedqualityDmaintenanceEnables AI assistants to programmatically edit, analyze, and export audio projects through MCP tools, including multi-track editing, effects, transcription, and semantic search.1-
- FlicenseNot gradedqualityCmaintenanceEnables AI clients like Claude and ChatGPT to generate images and videos, animate images, create lip-synced videos, list TTS voices, and manage media via remote MCP tools.-
- AlicenseAqualityBmaintenanceEnables MCP clients to run hosted audio tools such as stem splitting, analysis, segmentation, mastering, sample-pack generation, audio-to-MIDI conversion, music generation, synth-preset creation, and text-to-vox singing.19387 PyPIMIT