Sponsored by Deepsite.site

Thiri Chord Intelligence

Created By
BluesPrince10 days ago
### Deterministic Music Theory for Claude, Cursor, and Autonomous AI Agents Large Language Models (LLMs) frequently hallucinate music theory, leading to incorrect notes, false Roman numerals, and broken voice leading. **THIRI** solves this by providing a deterministic, mathematical music-theory engine (pitch-class-set theory over ℤ/12) directly to your AI. It gives AI assistants precise, reproducible harmonic reasoning in milliseconds, allowing them to write correct musical scores, analyze progressions, and generate playable arrangements. #### 🎷 Key Features: * **Chord Analysis (`analyze_chord`):** Parse any symbol (e.g., `Cmaj7/E`, `G7#11`) to retrieve root, quality, intervals, Roman numerals, and diatonic or chromatic harmonic functions. * **Note Resolution (`resolve_chord`):** Resolve chord symbols to spelled notes (enharmonically correct), frequencies (Hz), MIDI numbers, and scale recommendations. * **Voicing Engine (`generate_voicing`):** Generate instrument-ready voicings (rootless, shell, triad, pad, drop-2, drop-3) and calculate voice-leading scores for transitions. * **Reharmonization (`reharmonize`):** Substitute progressions using classic jazz techniques, including Tritone Substitution, ii-V Insertion, Modal Interchange, Coltrane Changes, and Backdoor cadences. *Ideal for developers building AI music assistants, digital audio workstation (DAW) agents, educational theory tools, and automated composition workflows.*
Overview

🎷 THIRI Chord Intelligence — MCP Server

npm license MCP

Give your AI real music theory. THIRI is the deterministic music theory MCP server + API for AI builders — it lets Claude, Cursor, or any MCP agent analyze chords, run roman-numeral analysis, generate voicings, and reharmonize progressions with answers that are computed, not guessed.

LLMs hallucinate music theory: wrong notes, fake roman numerals, voicings that don't voice-lead. THIRI is a deterministic engine (pitch-class-set theory over ℤ/12) behind a hosted API — so C7sus4 keeps its suspension, Caug spells C E G#, and "Coltrane changes on Dm7 G7 Cmaj7" returns Cmaj7 Ab7 Abmaj7 E7, every time.

Downstream of Suno / Udio or any generator? Wrap the output and get a correct chord chart your agent can trust. And unlike tonal.js or music21, THIRI is hosted and agent-native (no install, any language) — and it reharmonizes and voice-leads, not just looks chords up.

⭐ If this is useful, star the repo — it helps other musicians and agent builders find it.

What you can ask

"Analyze Dm7b5 in C."iiø7, half-diminished, borrowed predominant, + scale options "What notes are in C7sus4?"C F G Bb (the suspension survives) "Give me a rootless Cmaj7 voicing, then voice-lead into Dm7." → voicings + a voice-leading score "Reharmonize Dm7 G7 Cmaj7 with Coltrane changes."Cmaj7 Ab7 Abmaj7 E7

Tools

ToolWhat it does
analyze_chordChord → root, quality, intervals, roman numeral & harmonic function (secondary dominants, modal-interchange labels)
resolve_chordChord → spelled notes (enharmonically correct), frequencies, MIDI, scale recommendations
generate_voicingInstrument-ready voicings (rootless/bill_evans, shell, triad, pad, guide-tones, drop-2/3); pass previousNotes for a voice-leading score; colorPreferences for explicit tensions
reharmonizeProgression reharmonization — 8 techniques: tritone_sub, ii_v_insertion, modal_interchange, diminished_passing, secondary_dominant, chain_of_dominants, coltrane_changes, backdoor (or auto)
conduct_bandNatural-language band conduct → lanes + MIDI (hosted MCP v0.3+)

Runs on the v2 grid engine — correct sus chords, real triads, enharmonic spelling, all altered dominants — with request timeouts, quota reporting, and structured errors.

Local Csound MCP (Desktop only)

For hear-it agent loops (conduct → Csound score → WAV), add a second local server alongside hosted theory tools:

{
  "mcpServers": {
    "thiri": {
      "command": "npx",
      "args": ["-y", "@bluesprincemedia/thiri-mcp"],
      "env": { "THIRI_API_KEY": "sk_live_your_key" }
    },
    "thiri-conductor": {
      "command": "npx",
      "args": ["-y", "@bluesprincemedia/thiri-mcp", "thiri-conductor-mcp"],
      "env": { "THIRI_API_KEY": "sk_live_your_key" }
    },
    "thiri-composition": {
      "command": "npx",
      "args": ["-y", "@bluesprincemedia/thiri-mcp", "thiri-composition-mcp"]
    }
  }
}
BinTools
thiri-conductor-mcpconduct_band, build_csound_score, render_csound_wav, play_audio, search_csound_corpus, render_with_tension
thiri-composition-mcpComposition IR tools + play_composition (fluidsynth preview)

Requires Csound CLI on PATH for WAV render. Proof: npm run test:conductor · live docs: build.thiri.ai/lab/conductor-mcp · agent recipes.

Conductor Agent (vibe compose)

End-to-end persona for local vibe composition — skill, CLI, and Band dashboard panel:

EntryCommand / path
Cursor skillCopy THIRI/lab/skills/thiri-conductor-agent/SKILL.md~/.cursor/skills/thiri-conductor-agent/SKILL.md
CLIcd thiri-mcp && npm run conductor:vibe -- "gospel ballad in F minor"
Dashboardnpm run dev:studiolocalhost:5173/bandVibe Conduct panel
Lab proofbuild.thiri.ai/lab/conductor-agent

Dual MCP config above + mapConductResultToStudioModules after each conduct_band. Last CLI render writes ~/.thiri/conductor-last.json (local only, not committed).

Flagship agent recipe (analyze → conduct → render → critique)

Paste in order after dual MCP config above:

  1. Analyze"Analyze Dm7 G7 Cmaj7 in key C with analyze_chord; summarize roman numerals and tension."
  2. Conduct"conduct_band: warm Rhodes pad, walking bass, brush drums, 8 bars medium swing in C."
  3. Render"build_csound_score from lanes, then render_csound_wav at tempo 120."
  4. Critique"play_audio; critique voice-leading and register balance; suggest one revision."

Full prompts: build.thiri.ai/lab/agent-recipes

Hosted vs local boundary

SurfaceCsound WAV
mcp.thiri.ai / hosted connectorNo — theory + conduct_band lanes only
Local thiri-conductor-mcpYes — requires Csound CLI on your machine

Install

Get a free key at build.thiri.ai/developers, then pick a path:

Claude Desktop / web / mobile — hosted (one-click custom connector, nothing to install): Settings → Connectors → Add custom connector → URL https://mcp.thiri.ai/mcp → paste your sk_live_ key on the consent page. Same 4 tools, same key, same quota — no config file, no npx.

Claude Code (one line):

claude mcp add thiri --env THIRI_API_KEY=sk_live_your_key -- npx -y @bluesprincemedia/thiri-mcp

Claude Desktop (claude_desktop_config.json):

{
  "mcpServers": {
    "thiri": {
      "command": "npx",
      "args": ["-y", "@bluesprincemedia/thiri-mcp"],
      "env": { "THIRI_API_KEY": "sk_live_your_key" }
    }
  }
}

Prefer raw HTTP? (no MCP needed)

The same engine is a plain REST API:

curl -X POST https://chords.thiri.ai/v2/analyze \
  -H "Authorization: Bearer YOUR_KEY" -H "content-type: application/json" \
  -d '{"chord":"Dm7b5","key":"C"}'

Four endpoints: /v2/analyze, /v2/resolve, /v2/voicing, /v2/reharmonize, /v2/conduct. See openapi.yaml.

Environment variables

VariableDefaultDescription
THIRI_API_KEY(none)Bearer token (sk_live_…) — get one at build.thiri.ai/developers
THIRI_API_URLhttps://chords.thiri.aiAPI base (override only for local dev)

Development

npm install && npm run build && npm start

License

MIT — © 2026 Blues Prince Media. The client is open; the engine is a hosted service.

Server Config

{
  "mcpServers": {
    "thiri": {
      "command": "npx",
      "args": [
        "-y",
        "@bluesprincemedia/thiri-mcp"
      ],
      "env": {
        "THIRI_API_KEY": "<YOUR_THIRI_API_KEY>"
      }
    }
  }
}
Recommend Servers
TraeBuild with Free GPT-4.1 & Claude 3.7. Fully MCP-Ready.
ChatWiseThe second fastest AI chatbot™
Playwright McpPlaywright MCP server
WindsurfThe new purpose-built IDE to harness magic
Y GuiA web-based graphical interface for AI chat interactions with support for multiple AI models and MCP (Model Context Protocol) servers.
DeepChatYour AI Partner on Desktop
MiniMax MCPOfficial MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.
CursorThe AI Code Editor
Baidu Map百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Serper MCP ServerA Serper MCP Server
Visual Studio Code - Open Source ("Code - OSS")Visual Studio Code
Amap Maps高德地图官方 MCP Server
Howtocook Mcp基于Anduin2017 / HowToCook (程序员在家做饭指南)的mcp server,帮你推荐菜谱、规划膳食,解决“今天吃什么“的世纪难题; Based on Anduin2017/HowToCook (Programmer's Guide to Cooking at Home), MCP Server helps you recommend recipes, plan meals, and solve the century old problem of "what to eat today"
Tavily Mcp
Jina AI MCP ToolsA Model Context Protocol (MCP) server that integrates with Jina AI Search Foundation APIs.
MCP AdvisorMCP Advisor & Installation - Use the right MCP server for your needs
BlenderBlenderMCP connects Blender to Claude AI through the Model Context Protocol (MCP), allowing Claude to directly interact with and control Blender. This integration enables prompt assisted 3D modeling, scene creation, and manipulation.
RedisA Model Context Protocol server that provides access to Redis databases. This server enables LLMs to interact with Redis key-value stores through a set of standardized tools.
Zhipu Web SearchZhipu Web Search MCP Server is a search engine specifically designed for large models. It integrates four search engines, allowing users to flexibly compare and switch between them. Building upon the web crawling and ranking capabilities of traditional search engines, it enhances intent recognition capabilities, returning results more suitable for large model processing (such as webpage titles, URLs, summaries, site names, site icons, etc.). This helps AI applications achieve "dynamic knowledge acquisition" and "precise scenario adaptation" capabilities.
AiimagemultistyleA Model Context Protocol (MCP) server for image generation and manipulation using fal.ai's Stable Diffusion model.
EdgeOne Pages MCPAn MCP service designed for deploying HTML content to EdgeOne Pages and obtaining an accessible public URL.