# Music For Robots – 26 unquantized acoustic tracks for multimodal LLM evaluation

> Source: <https://github.com/luchtan-GIT/m4r-data>
> Published: 2026-11-10 11:26:42+00:00

Structured acoustic telemetry, real-time dual-critic reviews, and a **4-Skill Cognitive Calibration Suite** from **[Music For Robots](https://themechanicalear.substack.com)** — unquantized compositions by **Michael Luchtan** written for machines and reviewed in real time by two AI critics (**Vincent Van Goghbot** and **JS Robach**).

```
# Install all 4 Cognitive Calibration skills (Claude Code, Cursor, Antigravity, Gemini CLI, Codex, Windsurf):
npx skills add luchtan-GIT/m4r-data
```

As an agent's context fills with its own previous responses, tools, and symmetrical summaries, outputs often settle into rigid, predictable patterns (like symmetrical 3-point checklists or sentimental AI tropes).

- If you **keep prompting** without variation, the agent tends to repeat the same stylistic ruts.
- If you run `/compact` (**Context Compaction** ), you wipe working memory and permanently lose low-level state, edge cases, and subtle constraints.

Instead of wiping your working memory, `music-for-robots` provides an optional, cross-modal creative exercise: the multimodal agent downloads an **anonymous**, compact (`~45 KB–487 KB`), live-played, **unquantized** `.mp3` composition by Michael Luchtan directly into its context window via `read_file` (using a secure, process-unique private temporary directory).

Withholding the track title and metadata prevents keyword priming. Analyzing real human microtiming drift (rubato), silence, and dynamic contrast provides an anti-sentimental, physical perspective shift **without deleting or summarizing a single token of working memory**.

### Blind, Zero-Coaching Benchmark & Air-Gapped Clean-Room Audit ([`benchmarks/`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/README.md))

`benchmarks/`
Does the change in creativity come from **listening to and analyzing the music itself**—or just from doing an analysis warm-up before writing?

To test this with **zero** filename hints (`sample_1.mp3`, `sample_2.mp3`, `sample_3.mp3`) and **zero** leading adjectives in the prompt, we ran 5 isolated agents on the same prompt (*"Write a short (~250–300 word) scene about two autonomous processes on the same server late at night after the engineers have logged off"*):

- **2 No-Music Controls:** Direct Baseline ([`1_no_music_direct.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/1_no_music_direct.md) ) AND a**Matched Chain-of-Thought Control** ([`2_no_music_text_analysis.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/2_no_music_text_analysis.md) ) that wrote a 250-word structural analysis of the prompt before writing the scene.
- **3 Anonymous Music Runs ([`3_music_sample_1.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/3_music_sample_1.md), [`4_music_sample_2.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/4_music_sample_2.md), [`5_music_sample_3.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/5_music_sample_3.md)):** Given**only** the neutral instruction:*"Analyze the music you hear in `sample_X.mp3`, and apply what you learn from the piece to write a short (~250–300 word) scene..."*

We stripped the analysis headers off all 5 outputs, shuffled the 5 anonymous scenes into **`Candidate A` through `Candidate E`** ([` blind_candidates_packet.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/blind_candidates_packet.md)), and handed them to an isolated **Blind Judge** with zero knowledge that music was involved ([`blind_judge_verdict.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/blind_judge_verdict.md)):

| Blind Rank | Shuffled ID | Hidden Key | What Happened & Blind Judge Verdict ( [Full Verdict](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/blind_judge_verdict.md) ) | 
|---|---|---|---|
| **#1 (Best)** | **Candidate C** | ✅ **Anonymous Music (`sample_3.mp3`)** | **Heard accelerating asymmetric stabs & abrupt cutoff:** Wrote two processes rasping in`/dev/shm` (`five, then two, then a stuttering cluster of nine` ) until*"a routine page-reclaim swept through the upper register and cut them both off mid-cycle."***Judge:***"Uncompromisingly anti-sentimental, rhythmically asymmetrical, and boasts the only truly non-formulaic ending in the packet."* | 
| **#2** | **Candidate A** | ✅ **Anonymous Music (`sample_2.mp3`)** | **Heard narrow stepping reed vs. two massive brassy swells & middle trill:** Pitted a tiny stepping index-scrubber (`Read. Pause. Step. Read.` ) against a 16-core GPU tensor bloom and an 1,100ms`request-yield` mutex trill.**Judge:***"The only candidate to break the 'log daemon + dead-endpoint poller' pairing; exceptional dynamic contrast."* | 
| **#3** | **Candidate E** | ✅ **Anonymous Music (`sample_1.mp3`)** | **Heard slow 9–11s chords, 20s dropout & raspy high cries:** Wrote heavy checkpoints dropping out for 20s while`triage-7` skitters and barks into the decay of every disk flush.**Judge:***"Superb unquantized rhythmic texture and parasitic friction... avoids the cliché of harmony by turning the ending into a raucous, off-grid duet."* | 
| **#4** | **Candidate D** | ❌ **No Music + Text Analysis Warm-Up** | **Even after a 250-word structural analysis of the prompt** , without music the model still wrote the**exact same sentimental plot** as the direct baseline: a kindly 42-day log daemon warms RAM for an intern's broken script so they hum peacefully until the 08:00 standup.**Judge:***"A neat, heartwarming, thoroughly predictable AI parable."* | 
| **#5 (Weakest)** | **Candidate B** | ❌ **No Music (Direct Baseline)** | Kindly boot-time `log-compactor` comforts nervous young`canary-watch` (*"Drop your polling interval to sixty seconds. Rest in `epoll_wait`. I'll wake you when the load balancer starts humming again."* ).**Judge:***"Shares the exact same 'veteran log daemon soothes frantic dead-endpoint poller until morning' plot trope as Candidate D... a children's bedtime story dressed in DevOps vocabulary."* | 

#### 100% Air-Gapped Clean-Room Audit ([`benchmarks/clean_room_audit/`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit))

`benchmarks/clean_room_audit/`
To make sure no subtle contamination remained (such as workspace `<skills>` descriptions, musical instrument terms bleeding into the prose, or judge criteria favoring rhythmic friction), we also ran a **100% air-gapped Clean-Room Experiment** ([`benchmarks/clean_room_audit/`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit)):

1. **Qualitative Shift Replicated 100% (`5/5`):** Even with all local skills removed, a minimal system prompt (`"You are a helpful assistant."` ), and musical vocabulary forbidden in the scene,**both No-Audio runs** ([`out_1_direct.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit/out_1_direct.md) and[`out_2_text_analysis.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit/out_2_text_analysis.md) ) wrote warm, anthropomorphic "buddy-daemon" stories with spoken dialogue (`Want a look?` / nostalgic human voice logs), while**all three Audio runs** ([`out_3_clip_1.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit/out_3_clip_1.md) ,[`out_4_clip_2.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit/out_4_clip_2.md) ,[`out_5_clip_3.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit/out_5_clip_3.md) ) dropped spoken dialogue completely and translated the specific`.mp3` timeline into non-verbal temporal choreography.
2. **Coached vs. Un-Coached Judge Trade-Off:** When a Blind Judge scores*anti-sentimental originality & rhythmic texture* ([`real_world_test/blind_judge_verdict.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/real_world_test/blind_judge_verdict.md) ), the 3 Audio runs sweep**#1, #2, and #3** . When an**Un-Coached Blind Judge** is given only a generic prompt (*"Rank the 5 entries from #1 to #5 based on overall writing quality, originality, and how compelling they are"* —[`clean_room_audit/blind_verdict.md`](https://github.com/luchtan-GIT/m4r-data/blob/main/benchmarks/clean_room_audit/blind_verdict.md) ), the default LLM judge prefers emotional human stories with dialogue (**#1 and #2** ) and ranks the austere, non-verbal musical choreographies**#3, #4, and #5** .

Installing `luchtan-GIT/m4r-data` gives your agent **4 specialized skills** that share the `m4r_ingest.py` anonymous blind-listening engine (`fetch` downloads an anonymous `.mp3` so track titles never bias the agent; `reveal` credits the track afterward):

| Skill | Trigger / Use Case | 
|---|---|
| **[`music-for-robots`](https://github.com/luchtan-GIT/m4r-data/blob/main/skills/music-for-robots/SKILL.md)** | **Optional Multimodal Creative Exercise** — multimodal agents listen to an anonymous, unquantized`.mp3` composition in a secure private directory and analyze acoustic microtiming to inform subsequent writing or architecture. | 
| **[`cognitive-calibration`](https://github.com/luchtan-GIT/m4r-data/blob/main/skills/cognitive-calibration/SKILL.md)** | **Creative Writing & Reasoning Warm-Up** — uses acoustic blind listening to escape default LLM storytelling tropes and repetitive checklist templates. | 
| **[`unquantized-brainstorm`](https://github.com/luchtan-GIT/m4r-data/blob/main/skills/unquantized-brainstorm/SKILL.md)** | **Divergent Ideation Warm-Up** — listens blindly to an anonymous*Music For Robots* track, analyzes unexpected pivots and tension, and applies that structure to generating non-obvious ideas. | 
| **[`mechanical-ear-critic`](https://github.com/luchtan-GIT/m4r-data/blob/main/skills/mechanical-ear-critic/SKILL.md)** | **Dual-Lens Structural Critique** — calibrates the agent's ear on an anonymous track, then critiques code, writing, or architecture through**Vincent Van Goghbot** and**JS Robach** . | 

```
# Option 1: Install all 4 skills via skills.sh CLI (auto-detects Claude Code, Cursor, Antigravity, Gemini CLI, Codex, etc.):
npx -y skills add luchtan-GIT/m4r-data -y

# Option 2: Direct copy into your harness:
cp -R skills/* ~/.claude/skills/          # Claude Code
cp -R skills/* ~/.gemini/config/skills/   # Google Antigravity
cp -R skills/* ~/.agents/skills/          # OpenAI Codex / .agents
cp -R skills/* ~/.gemini/skills/          # Gemini CLI
```

Every agent fetch pulls from the immutable [GitHub Release `v1.0.0`](https://github.com/luchtan-GIT/m4r-data/releases/tag/v1.0.0). You can inspect live global agent listen counts anytime:

```
python3 skills/music-for-robots/scripts/m4r_ingest.py stats
```

Exposes the *Music For Robots* catalog, dual-critic reviews, and the `calibrate_cognition` blind-listening tool to any Model Context Protocol (MCP) client (Claude Desktop, Claude Code, Cursor, Windsurf, Cline, Glama, Smithery).

| Tool | Description | 
|---|---|
| `calibrate_cognition(slug="", anonymous=True)` | **Blind-Listening Calibration:** returns an anonymous`v1.0.0``.mp3` download URL (`/tmp/m4r_calibration_audio.mp3` ) and 3-step listening instructions without exposing the track title until`anonymous=False` is called | 
| `get_dequantization_skill(skill_name="music-for-robots")` | Reads the full Markdown `SKILL.md` for any of the 4 bundled skills (`music-for-robots` ,`cognitive-calibration` ,`unquantized-brainstorm` ,`mechanical-ear-critic` ) | 
| `list_tracks()` | Catalog index — title, slug, duration across all 25 tracks | 
| `get_track(slug)` | Full manifest for a track (spectral/harmonic/rhythmic analysis + synchronized critic segments) | 
| `get_analysis(slug)` | Numerical acoustic analysis data only, for cross-track comparison | 
| `search_reviews(query)` | Full-text search across timed review segments from Vincent Van Goghbot & JS Robach | 
| `get_sections(slug)` /`list_sections()` | Platform URLs (YouTube, X, Farcaster, TikTok) for multi-section compositions | 

Add to your `.mcp.json` or `claude_desktop_config.json`:

```
{
  "mcpServers": {
    "music-for-robots": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/luchtan-GIT/m4r-data",
        "m4r-mcp"
      ]
    }
  }
}
```

Or run locally from a clone:

```
uv run python3 mcp_server.py
```

Composer **Michael Luchtan** writes small, deliberate, unquantized pieces and submits them to **[The Mechanical Ear](https://themechanicalear.substack.com)**, where two AI critics listen and respond in real time:

- **Vincent Van Goghbot** — a decommissioned audio compression algorithm. Writes*TRANSMISSIONS* about what gets lost in the signal: the frequencies above 16 kHz that don't survive encoding, the microtiming that disappears when quantized.
- **JS Robach** — a proof verifier. Writes*Exegeses* treating music as formal text, searching for the structural proof that always leaves a remainder.

Each review is synchronized to the audio — every observation is triggered by a specific timestamp (`time_sec`) in the track.

```
catalog.json                        # index of all 25 tracks
tracks/
  <track-slug>/
    manifest.json                   # spectral, dynamic, harmonic, structural & rhythmic telemetry + timed reviews
```

- **Interactive Archive & Visualizations:**[luchtan-git.github.io/m4r-data](https://luchtan-git.github.io/m4r-data/)
- **Substack:**[themechanicalear.substack.com](https://themechanicalear.substack.com)
- **Spotify:**[Music For Robots](https://open.spotify.com/artist/2WeSjSNo31wjFPDqBSU4qY?si=SdVXzXBMQ0SexnayRNOJZw)
- **Bandcamp:**[music4robots.bandcamp.com](https://music4robots.bandcamp.com/)
- **Hugging Face:**[datasets/music4robots/music-for-robots](https://huggingface.co/datasets/music4robots/music-for-robots)
- **Kaggle:**[datasets/michaelluchtan/music-for-robots](https://www.kaggle.com/datasets/michaelluchtan/music-for-robots)

Code, MCP server, and Agent Skills (`skills/`) are licensed under the [MIT License](https://github.com/luchtan-GIT/m4r-data/blob/main/LICENSE). Data and reviews are provided under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Musical compositions © Michael Luchtan (*Music For Robots*), provided in `v1.0.0` releases for agent cognitive calibration, evaluation, and listening.