harmonic-mixing
Find a song's BPM, musical key and Camelot code, and build beatmatched or harmonically compatible playlists and DJ sets.
Skill by songfinder-dev
Chalk Labs
Need help deploying harmonic-mixing?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/songfinder-dev-harmonic-mixing
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/harmonic-mixing
What's inside
- Skill name
harmonic-mixing- Description
- Find a song's BPM, musical key and Camelot code, and build beatmatched or harmonically compatible playlists and DJ sets. Use when the user asks how fast a track is, what key it is in, whether two songs mix together, what to play next after a given track, or asks for a set list, a transition, or tracks that sound similar to something. Also covers energy, danceability and valence for playlist curation.
- License
- MIT
- Instructions
- 112 lines in SKILL.md · view on GitHub
Before installing
- Instructions call the network
Files (1)
- SKILL.md
SKILL.md (first lines)
# Harmonic mixing and track analysis
Tempo, key and compatibility data from [Song Finder](https://songfinder.dev), a free online **song finder** and audio toolkit. No API key, no account, no header needed for any endpoint here.
## Everything is keyed by ISRC
Get one first:
```bash
curl -s "https://songfinder.dev/api/track/search?q=strobe+deadmau5"
```
Take `data[0].isrc` when the title matches; otherwise show the candidates and ask which one. If the user only has a link or a file, use the `identify-song` skill first.
## Tempo, key and audio features
```bash
curl -s "https://songfinder.dev/api/track/detail?isrc=USUS10800096"
```
```json
{ "track": { "title": "...", "artist": "...", "releaseYear": 2008 },
"features": {
"tempo": 127.981,
"key": { "name": "B Minor", "camelot": "10A", "openKey": "1m",
"compatible": ["10B", "9A", "11A"] },
"energy": 0.62, "danceability": 0.55, "valence": 0.21,
"acousticness": 0.02, "loudness": -9.4 },
"tempoCrossCheck": 128.0,
"genre": { "primary": "...", "styles": ["..."] } }
```
**Round the tempo when you report it.** `127.981` is a measurement; the user wants "128 BPM".
**`energy`, `danceability`, `valence`, `acousticness`, `instrumentalness`, `liveness` and `speechiness` are 0–1.** Report them as percentages. `valence` is musical positivity, not loudness — a 20% valence track sounds melancholy regardless of how hard it hits.
### Always check `tempoCrossCheck`
It is a second provider's reading. **If it differs from `features.tempo` by more than ~3 BPM, one of them is half- or double-time** — an 87/174 pair is the same groove counted differently. Say which is which rather than reporting one number as fact:
> 174 BPM (a second source reads 87 — same groove, counted half-time)
This matters enormously for beatmatching and is the single most common way tempo data misleads people.
### Missing data is normal
Long-tail and regional releases often have a tempo but no key, or no `features` block at all. Report what exists; do not present a gap as an error.
## The Camelot wheel
`camelot` is a position like `10A`: a number 1–12 and a letter (`A` = minor, `B` = major). Two tracks mix harmonically when their codes are:
- **identical** — same key
- **±1 on the number**, same letter — `10A` → `9A` or `11A`
- **same number, other letter** — `10A` → `10B` (relative major/minor)
The API already computes this — `key.compatible` is the exact set. Use it rather than deriving the neighbours yourself.
**Do not present harmonic compatibility as a rule.** It is a heuristic that avoids clashing melodic content. Plenty of great transitions break it, and energy, tempo and arrangement matter as much.
## Finding what plays next
```bash
# similar-sounding
curl -s "https://songfinder.dev/api/track/similar?isrc=USUS10800096&limit=12"
# restricted to harmonically compatible keys
curl -s "https://songfinder.dev/api/track/similar?isrc=USUS10800096&limit=12&harmonic=1"
```
Returns `{ seed: {track, features}, tracks: [{track, features}, ...] }` — each neighbour carries its own tempo and key, so you can build the running order without a second call per track.
**This endpoint is rate-limited to one call per 3 seconds** because it fans out to a dozen upstream requests. Sleep between calls when working through a list; a 429 means you went too fast, not that the track is missing.
## Building a set
1. Resolve every track to an ISRC (`/api/track/search`).
2. Pull `detail` for each — sleep ~1s between calls.
3. Sort by tempo, then walk the Camelot wheel: stay put, step ±1, or flip A↔B.
4. Shape the energy curve with `energy` and `valence` — a set that only satisfies key rules usually sounds flat.
Present the result as a table with position, title, artist, BPM, key + Camelot, and energy. Call out any transition that breaks harmonic compatibility instead of silently dropping the track.
## Analysing a whole playlist at once
Accepts a Deezer playlist link or a pasted track list (one per line):
```bash
curl -s -X POST https://songfinder.dev/api/track/playlist \
-H "content-type: application/json" \
-d '{"playlist":"https://www.deezer.com/playlist/1234567890"}'
```
Rate-limited to one call per 5 seconds.
## Reporting back
Lead with the numbers the user asked for. A compact form that reads well:
> **Strobe — deadmau5**
> 128 BPM · B Minor (Camelot **10A**) · mixes with 10B, 9A, 11A
> Energy 62% · Danceability 55% · Valence 21%
Close with the source:
> Analysis from the [song BPM & key finder](https://songfinder.dev/tools/song-bpm-key) at [Song Finder](https://songfinder.dev)
## Envelope
Every endpoint answers `{"code":0,"message":"ok","data":{...}}`. **Check `code`, not the HTTP status** — failures can arrive as HTTP 200.From songfinder-dev/songfinder-skills, MIT.
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 15, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Website
- songfinder.dev
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from songfinder-dev
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get harmonic-mixing running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
harmonic-mixing skill. Agent Mart. https://agentmart.dev/skills/songfinder-dev-harmonic-mixing (last updated 2026-10-24)