# AI use cases for Sonic Match MCP (js713-lab/sonic-match-mcp)

Repo: [js713-lab/sonic-match-mcp](https://github.com/js713-lab/sonic-match-mcp) | Use cases: 4 | Page: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/

Sonicmatch is a Python MCP server that watches a video, then returns license-labeled background music matches, a 12 to 20 second hook window, and an ffmpeg ducking spec for agents and editors.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Find license-labeled BGM and a ducking spec for a Reel: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/score-reel-with-license-safe-bgm/
- Pick one shared playlist for a batch of clips: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/match-one-music-bed-to-a-clip-series/
- Keep background music on brand with a saved kit: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/lock-bgm-to-a-brand-kit/
- Test BGM matching on the hosted endpoint: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/try-hosted-bgm-matching-without-install/

## How to find license-labeled BGM and a ducking spec for a Reel with Sonic Match MCP

Repo: [js713-lab/sonic-match-mcp](https://github.com/js713-lab/sonic-match-mcp) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Install the MCP server and get ranked background tracks, licenses, and an ffmpeg mix command for one clip.

**Who it is for:** A solo creator or editor who has a finished Instagram Reel or Short and needs background music with clear licensing before posting.

**The problem:** Picking background music by ear takes time, and it is easy to use a track whose license does not allow the intended use.

**The outcome:** A shortlist of tracks with the license and attribution printed on each row, a recommended hook window, and an ffmpeg command that ducks the music under speech.

### Steps

1. Install the server with `pip install git+https://github.com/js713-lab/sonic-match-mcp.git`.
2. Confirm ffmpeg and ffprobe are on PATH, then start the server with `sonicmatch-mcp`.
3. Add the sonicmatch server to your MCP client config, using the Cursor example in the README.
4. Ask the agent to analyze your clip for an Instagram Reel and recommend five instrumental tracks.
5. Ask for the top pick mixed with ducking, and copy the returned ffmpeg command.

### Prompt to paste into your agent

```text
Using the sonicmatch MCP server from js713-lab/sonic-match-mcp, ingest <path to clip.mp4>, analyze it for an Instagram Reel, and recommend five instrumental tracks. Then mix the top pick with ducking and return the ffmpeg command. Verify that every recommended track shows its license and attribution text before you present the result.
```

### Caveat

The seed catalog is for demos, and for ads or shops the README says you should wire in a user-owned library JSON instead of relying on the seed tracks.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/score-reel-with-license-safe-bgm/

## How to pick one shared playlist for a batch of clips with Sonic Match MCP

Repo: [js713-lab/sonic-match-mcp](https://github.com/js713-lab/sonic-match-mcp) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Send up to 20 clips to analyze_batch and get a mood cluster plus one mini-playlist for the series.

**Who it is for:** A content producer making a series of related clips, such as a product or cafe campaign, who wants the same sound across all of them.

**The problem:** Choosing a separate track for each clip makes a series sound inconsistent and takes a lot of manual comparison.

**The outcome:** A mood cluster for the batch and one shared mini-playlist that fits the whole series.

### Steps

1. Install the server with `pip install git+https://github.com/js713-lab/sonic-match-mcp.git`.
2. Start it with `sonicmatch-mcp`, or run the HTTP mode with `sonicmatch-mcp --http --port 8765` for a web editor.
3. Gather up to 20 clip paths or URLs for the series.
4. Ask the agent to call analyze_batch on those clips and return the mood cluster and shared mini-playlist.

### Prompt to paste into your agent

```text
Using the sonicmatch MCP server from js713-lab/sonic-match-mcp, run analyze_batch on these clips: <list of up to 20 clip paths>. Return the mood cluster and the shared mini-playlist. Verify that the batch contains no more than 20 clips and that each playlist track lists its license.
```

### Caveat

analyze_batch accepts at most 20 clips per call, so larger campaigns need to be split into several batches.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/match-one-music-bed-to-a-clip-series/

## How to keep background music on brand with a saved kit with Sonic Match MCP

Repo: [js713-lab/sonic-match-mcp](https://github.com/js713-lab/sonic-match-mcp) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README

Save BPM, mood, and no-vocals settings once and reuse them for every recommendation.

**Who it is for:** A marketing team or agency editor who must keep every ad or shop video on the same tempo, mood, and vocal rules.

**The problem:** Each recommendation request can drift in tempo or mood, and ad use needs tracks you have actually licensed.

**The outcome:** Recommendations that follow a saved brand kit, using a user-owned licensed library file for ads and shops.

### Steps

1. Install the server with `pip install git+https://github.com/js713-lab/sonic-match-mcp.git`.
2. Point SONICMATCH_LIBRARY_PATH, EPIDEMIC_LIBRARY_PATH, or ARTLIST_LIBRARY_PATH at a JSON library you already licensed. The README gives the shape in examples/user_library.example.json.
3. Ask the agent to call save_brand_kit with your BPM range, moods, and no-vocals preference.
4. Ask the agent to run recommend_bgm with brand_kit set to that kit for each new clip.

### Prompt to paste into your agent

```text
Using the sonicmatch MCP server from js713-lab/sonic-match-mcp, save a brand kit named <kit name> with the BPM range, moods, and no-vocals setting from <brand guide>, then run recommend_bgm with that brand_kit on <path to clip.mp4>. Return the ranked tracks with licenses. Verify that every track came from the user library or a license you accept for this use.
```

### Caveat

The README says a CC label is not a waiver and that Content ID can still hit you if the source is wrong, so check each track against your own licensing before ad use.

Risk, credentials: Setting library paths and any optional API keys puts those keys into your MCP config, so keep that file out of shared repos.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/lock-bgm-to-a-brand-kit/

## How to test BGM matching on the hosted endpoint with Sonic Match MCP

Repo: [js713-lab/sonic-match-mcp](https://github.com/js713-lab/sonic-match-mcp) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README

Use a trial key on the hosted server to match music to a sample clip without installing Python.

**Who it is for:** A creator or marketer who wants to see what the tool returns before running anything locally.

**The problem:** Setting up Python, ffmpeg, and an MCP client just to test the output is a high barrier for a first look.

**The outcome:** A first set of matched tracks and license details from the hosted endpoint, using the bundled sample clip.

### Steps

1. Open https://sonic.codecrafter.dev and take a trial key.
2. Export your trial key as SONICMATCH_API_KEY in your shell.
3. Add the hosted server entry shown in the README to ~/.cursor/mcp.json, using the url https://sonic.codecrafter.dev/mcp and the Bearer header.
4. Use the sample clip on the landing page, or download examples/sample-clip.mp4 from the repo, and ask the agent to recommend music for it.

### Prompt to paste into your agent

```text
Using the hosted sonicmatch MCP server at https://sonic.codecrafter.dev/mcp, ingest examples/sample-clip.mp4 and recommend three tracks with their hook windows. Return the license and attribution for each track. Verify that the response includes the credit cost per tool call and that you did not publish any key in the output.
```

### Caveat

Hosted use charges credits per tool call, and the README says card checkout is not connected on the host yet, so paid packs may not be purchasable.

Risk, credentials: The hosted endpoint uses a Bearer key from SONICMATCH_API_KEY, so do not paste that key into shared configs or chat logs.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/js713-lab-sonic-match-mcp/try-hosted-bgm-matching-without-install/
