openrouter-text2music

$npx mdskill add QinghongLin/data2story-skill/openrouter-text2music

Generate atmospheric sound-design or SFX via OpenRouter and Google Lyria 3 Pro.

  • Creates ambient beds or best-effort sharp foley when no real recording exists.
  • Uses OpenRouter API with Google Lyria 3 Pro model for 48kHz stereo audio.
  • Prompts must describe texture, not song elements like genre or melody.
  • Downloads generated audio to a specified project directory as a WAV file.

SKILL.md

.github/skills/openrouter-text2musicView on GitHub ↗
---
name: openrouter-text2music
description: Generate atmospheric sound-design / SFX (NOT speech, NOT the front-of-blog BGM) via OpenRouter using Google Lyria 3 Pro.
---

# openrouter-text2music

Text → audio via OpenRouter. Default model: `google/lyria-3-pro-preview`.

**Role in Data2Story: atmospheric sound-design / SFX only — never the front BGM.** Use this for a sound the story needs that has NO findable real recording: an ambient bed (tension drone, texture, room tone) or best-effort sharp foley (e.g. a glass-shatter texture). Lyria's strength is ambient/textural beds, not crisp transients, so sharp foley is best-effort. The result is an opt-in, event/section-bound, reader-controlled element. **The front-of-blog BGM must be a SOURCED real track (sourced_bgm via the Scout) — never generated here.**

**⚠️ This is a generative audio model, not TTS.** It produces 48kHz stereo audio; for narration/voiceover use a dedicated TTS tool (e.g., OpenAI `tts-1`, ElevenLabs).

## Usage

Resolve `TOOL_DIR` = the directory containing this `SKILL.md`. Commands below use `TOOL_DIR` as a symbolic placeholder; replace it with the resolved, quoted path before running Bash.

```bash
export OPENROUTER_API_KEY=sk-or-v1-...

python3 TOOL_DIR/scripts/generate_music.py \
  --prompt "Low rumbling tension drone, sub-bass swell, no melody, slowly building unease" \
  --download PROJECT_DIR/assets/sfx_drone.wav
```

## SFX prompting

Describe a TEXTURE, not a song — no genre/BPM/lead/vocals. Examples:

- `"low rumbling tension drone, sub-bass swell, no melody"`
- `"sharp glass-shatter texture, brittle high transients"` (foley — best-effort)
- `"distant wind and creaking-ice ambience, cold and hollow"`

## Flags

| Flag | Default | Description |
|---|---|---|
| `--prompt` | required | Text prompt — describe the atmospheric texture / SFX (mood, materials, motion); avoid song structure |
| `--download` | required | Output audio file path |
| `--model` | `google/lyria-3-pro-preview` | Alt: `google/lyria-3-clip-preview` (shorter clips) |

## Pricing

- `lyria-3-pro-preview`: $0.08 per call

## Notes

- Request uses `POST /api/v1/chat/completions` with `modalities: ["audio","text"]`.
- Response parsing handles several shapes: `message.audio`, a content part with `type=audio`, or a `data:audio/...` URL embedded in the text content.
- Output format is typically WAV (48kHz stereo); the script saves whatever bytes the API returns — choose the extension to match.

More from QinghongLin/data2story-skill

SkillDescription
analystExhaustively profile a dataset and list ALL possible analyses — distributions, correlations, rankings, trends, group comparisons, anomalies. Reads detective.json for context. Runs after the Detective/Scout, before the Editor. Outputs analyst.json with ana_xx IDs and chart-ready data_tables.
auditorAudit a generated Data2Story blog for build correctness across ALL modalities by ACTUALLY RENDERING it in a real headless browser (when available) — catching blank/0-width charts, broken/oversized media, desktop+mobile overflow, and dead interactions that source inspection cannot see — then having vision agents review the screenshots, playtesting every interactive, and walking the flagship capability contract. Plus VIEW every image for a wrong/AI-faked subject. Fixes layout in place; sends content-correctness problems back to the owning role. Use at Stage 6 after the Programmer assembles index.html and before the Critic; re-runs each revision round. Triggers: a built index.html exists, or you need to confirm charts/media/interactions actually render.
cinematographerDesign the MANDATORY cinematic scroll experience (every blog ships one): a full-bleed, scroll-driven background — primarily from the Scout's verified real imagery — plus the global motion choreography, so the narrative unfolds like a film as the reader scrolls. Owns the BACKGROUND layer + page-level motion (not the Designer's per-section visuals, not the Interaction centerpiece — it stages them inside the scroll). Outputs cinematographer.json (cin_xx scenes). There is no off path: a topic with no sourceable real imagery falls back to a generative-atmosphere or data-driven-spine background, never to a bare column.
copywriterName the piece — re-write the masthead (headline + standfirst + kicker), every section title, and every figure/photo/table caption to a research-driven titling standard, killing the AI-tell patterns (the 'flat statement. flat counter-statement.' two-beat above all) a competent default falls into. Reads editor.md/json + analyst.json + the resolved topic_profile; writes copywriter.json — STRINGS ONLY (masthead{headline,standfirst,kicker}, items{edt_xx:{title}, des_xx:{caption}}), each backed by a real ana_*. Names, never edits: it touches no finding, no number, no data-* id, no layout — so the Verify layer is untouched and the Programmer renders the masthead + figcaptions from copywriter.json verbatim. Runs at Stage 3.5, after the Editor, before the Designer.
criticReview a finished Data2Story blog against the 5 quality rubric dimensions (visual_design, narrative_pacing, data_method_transparency, claim_data_alignment, insight_value), score each 1-7 with on-page evidence, and emit critic.json with pass/fail + targeted, surgical send-back instructions. Verifies every load-bearing claim/asset against its traceability chain before scoring; applies the caveat-survival, honest-accuracy, and third-party-attribution caps. Does NOT rewrite content — scores and sends back. Use at Stage 6.5 after the Auditor and after verify.py has produced verifier.json; re-runs each revision round. Triggers: a built index.html plus verifier.json exist, or you need to judge whether the article is actually good.
data2storyData Journalist Agent (Data2Story) — orchestrator: turn a dataset into a blog. Runs detective → analyst → editor → designer → programmer → auditor → inspector in sequence. Creates a versioned project folder for each run.
data2story-proUse to turn a dataset into a verifiable multimedia blog (a data story / data-driven article / interactive dashboard from a dataset). Orchestrator for the Data Journalist Agent (Data2Story): a 7-team newsroom (14 agents) running detective → scout → analyst → imagineer → editor → copywriter → designer → interaction → hero → cinematographer → programmer → auditor → critic → inspector in sequence. Trigger when the user hands over a dataset (CSV/JSON/folder/path) and wants a published story, blog post, or interactive report built from it. Creates a versioned project folder per run.
dataviz-craftA shared reference library for editorial-grade data-visualization craft — Vega-Lite-first with a D3 fallback for charts Vega-Lite can't express. Read by the Designer at chart selection (intent → ranked chart type), the Programmer at implementation (editorial Vega-Lite recipes, annotation layers, axis/label de-clutter, encoding craft), and the Auditor/Critic for chart-quality review. It encodes the FT Visual Vocabulary intent taxonomy, the Cleveland–McGill channel-accuracy ordering, the BBC bbplot de-clutter ruleset, and colorblind-safe encoding rules. Not a pipeline stage — a craft source, like frontend-design.
designerRead editor.md, editor.json, and analyst.json. Make data-driven creative visual decisions for every section — teaser, charts, images, videos, audio, maps, and interactive demos when they fit the data. Generate selected assets. No HTML. Outputs designer.json with des_xx IDs.
detectiveResearch external context for a dataset — domain background, history, related studies, and why this data matters. Outputs detective.json (structured findings with det_xx IDs) before any analysis begins.