Stargazer Interpreter · 星解者

简体中文 · 繁體中文

An Agent Skills compatible skill that interprets Lenormand card readings from Stargazer's Oracle. Works with any AI agent that supports the Agent Skills open standard — not tied to a specific platform.

Paste a Stargazer prompt → get a full interpretation, saved as structured data, rendered into beautiful Kami-styled A4 PDFs, interactive web pages, or 1080×1440 social cards.

Design

Every product shares a unified visual language inspired by two sources:

• Kami — warm parchment background (#f5f4ed), ink-blue accent (#1B365D), Chiron Sung HK serif typography. Clean, literary, minimal.
• Geometric Silence — a custom 36-card Lenormand deck in architectural blueprint style. Muted sage green (#7A8B7A) linework on cream (#F5F0E8), with dusty rose (#C4A0A0) accents. Swiss International Style meets scientific specimen plates. No mystical ornament — pure geometric abstraction.

Together they create a reading experience that feels scholarly, warm, and quietly beautiful.

Features

• 6 spread types: 1-card Yes/No · 2-card pair · 3-card linear · 5-card linear · 9-card Box Spread (3×3) · A-or-B choice
• Bilingual: Auto-detects Chinese or English prompts; all output matches the prompt language
• Combination-first reading: Systematic combination methodology — adjacent card pair chains with modifier type system (Noun+Modifier model). Each card modifies its neighbor according to its modifier type; 8 negative modifiers automatically darken the tone. Neutral, honest, no empty reassurance.
• 4 product types from one reading: A4 PDF (default) · self-contained web page · 1080×1440 social card carousel · full-page PNG long image
• Persistent readings: Saved as structured JSON + human-readable Markdown; regenerate into any product format anytime
• Seed-based HTML: Copy + fill HTML skeletons — no fragile from-scratch construction
• Geometric Silence deck: 36 original SVG cards in architectural blueprint style
• Chiron font pair: Chiron Sung HK (昭源宋體) serif + Chiron Hei HK (昭源黑體) sans, loaded from Google Fonts CDN with CN mirror fallback

Preview

Product	Spread	Cards	Link
Live Demo	3-Card Daily	Paste → Prompt → Preview → Download	Demo
A4 PDF	5-Card	月亮 → 棺材 → 樹 → 雲 → 男人	PDF
A4 PDF	A-or-B (EN)	Ship·Bear vs Mountain·Ring·Book	PDF
Social Card	3-Card Daily	All 4 carousel pages	P1 · P2 · P3 · P4

Install

Tell your agent:

Install the stargazer-interpreter skill from GitHub: github.com/locoda/stargazer-interpreter

Or clone manually into your agent's skills directory:

git clone https://github.com/locoda/stargazer-interpreter.git ~/your-agent/skills/stargazer-interpreter/

Requires: Node.js + Puppeteer (auto-installs Chromium on first PDF/social card render).

Usage

First-time reading

Paste a Stargazer prompt — the agent runs Phase 1 (parse) + Phase 2 (interpret + save), generates A4 PDF by default.

User: Interpret this spread

You are a traditional Lenormand reader. Read for me...
---
Question: How will my career change in the next 3 months?
Date: 2026-06-20
Spread: 3-card linear
1. Clover
2. Child
3. Snake

Generate a different product

User: Generate social cards from the latest reading
User: Generate web page from reading #2

Request multiple products

User: Interpret this spread, generate PDF and social cards
[Stargazer prompt...]

Three-phase workflow

Stargazer Prompt
    │
    ▼
[Phase 1] Parse → structured card data
    │
    ▼
[Phase 2] Interpret → save reading (JSON + MD)
    │
    ▼
[Phase 3] Generate products from saved reading
           ├── A4 PDF (default)
           ├── Web Page
           ├── Social Card (1080×1440 carousel)
           └── Long Image (full-page PNG)

Structure

├── SKILL.md                         # Main instructions
├── README.md                        # English docs
├── README.zh-CN.md                  # Simplified Chinese docs
├── README.zh-TW.md                  # Traditional Chinese docs
├── assets/
│   ├── seed-a4-pdf.html             # A4 PDF HTML skeleton
│   ├── seed-social-card.html        # 1080×1440 social card skeleton
│   └── seed-web-page.html           # Web page skeleton
├── cards/
│   └── card-01-rider.svg ...        # 36 Geometric Silence SVG cards
├── references/
│   ├── lenormand-cards.md           # 36-card database (MIT), modifier types, methodology
│   ├── spread-parsing.md            # Prompt parsing rules
│   ├── design-system.md             # Shared fonts, colors, card slugs
│   ├── reading-schema.md            # JSON schema for persisted readings
│   └── template-a4-pdf.md           # A4 PDF chapter & page balance rules
└── scripts/
    ├── generate-pdf.js              # HTML → A4 PDF (Puppeteer)
    ├── render-social-cards.js       # HTML → 1080×1440 PNGs (Puppeteer)
    ├── render-long-image.js         # HTML → full-page PNG (Puppeteer)
    ├── validate.js                  # Product validation (static + Puppeteer)
    └── manage-readings.js           # Reading index CLI

Frontend & previews are maintained in the site branch and deployed via Cloudflare Pages at stargazer-interpreter.1mether.me. See the live demo.

License

MIT

Credits & References

Typography

• Chiron Sung HK (昭源宋體) — variable serif, SIL OFL 1.1
• Chiron Hei HK (昭源黑體) — variable sans, SIL OFL 1.1

Design

• Kami — visual style inspiration, MIT
• Geometric Silence — original card deck design

Data

• Stargazer's Oracle — online Lenormand platform
• lenormand-oracle — 36-card bilingual dataset, MIT
• Combination modifier system — original synthesis based on public-domain Lenormand tradition