---
name: BIRD Book Deconstructor
slug: bird-book-deconstructor
category: AI Engineering
description: BIRD Book Deconstructor converts complex manuscript text into BIRD 2.1 knowledge nodes with Book Address, Knowledge Index, Routes, and Deep Links. Use it to split text for TheBrain, Excel, Roam, Obsidian, or to audit and migrate existing BIRD indexes.
github: "https://github.com/twhsi/skills/tree/main/skills/thebrain-bird-address"
language: Python
stars: 259
forks: 38
install: "npx degit https://github.com/twhsi/skills/tree/main/skills/thebrain-bird-address ~/.claude/skills/thebrain-bird-address"
installs_to: ~/.claude/skills/thebrain-bird-address
source_path: skills/thebrain-bird-address/SKILL.md
collection_size: 19
category_size: 2631
collection_url: "https://dirskills.com/collections/twhsi/skills"
added: 2026-09-02T05:20:24.454Z
last_synced: 2026-09-02T05:20:24.454Z
canonical_url: "https://dirskills.com/skills/bird-book-deconstructor"
---

# BIRD Book Deconstructor

BIRD Book Deconstructor converts complex manuscript text into BIRD 2.1 knowledge nodes with Book Address, Knowledge Index, Routes, and Deep Links. Use it to split text for TheBrain, Excel, Roam, Obsidian, or to audit and migrate existing BIRD indexes.

**Install:**

```bash
npx degit https://github.com/twhsi/skills/tree/main/skills/thebrain-bird-address ~/.claude/skills/thebrain-bird-address
```

## README

# BIRD Book Deconstructor 2.2

Convert complex structural text into addressable, searchable, cross-linked book nodes for TheBrain, Excel, Roam Research, Obsidian, and AI Agents.

Version 2.2 is the Skill integration release. Canonical interchange objects remain `BIRD-2.1` for backward compatibility.

## Three Kings 2.2 Integration

BIRD is the **address and route king** in the three-skill publishing chain:

`iMandalArt 2.2 eight angles -> BIRD 2.2 Book Address + Index + Route + Deep Link -> A4 Booklet 2.2`

- Receive the center, eight angles, source identities, and spatial order from iMandalArt.
- Turn reusable claims into stable Book Addresses and `W + T + K + A` Knowledge Indexes.
- Send seven ordered Routes, citations, pending Deep Links, and merge reasons to the booklet manifest.

## Formal Protocol

- `B = Book Address`: stable book/project hierarchy address; answer "where in the book?"
- `I = Knowledge Index`: structured index object `W + T + K + A`; answer "what knowledge object?"
- `R = Route`: ordered cross-node or cross-chapter paths; answer "where does it lead?"
- `D = Deep Link`: exact permanent application address; answer "how is it opened?"

Read [`references/bird-2.1-spec.md`](references/bird-2.1-spec.md) before assigning Index Weight, Index Type, or Semantic Role.

Do not use the old `B = Branch` or unstructured `I = title` interpretation. Mark legacy input as `BIRD 1.x` and map it into 2.1. Preserve valid BIRD 2.0 fields during migration; add `Role` only when evidence supports it.

## Distinguish Two Types

- `StructuralType`: book function, such as `部 / 章 / 節 / 項`.
- `IndexType`: knowledge kind inside `I`, using `C / M / P / B / T / O / E / L / S / A / X`.

Never put `章` into `IndexType`, or `Concept` into `StructuralType`.

## Choose the Mode

- **Single Node**: analyze one existing Thought and output one BIRD object.
- **Deconstruct Text**: split complex prose into chapter, section, and atomic item nodes.
- **Excel Workbook**: create a validated `.xlsx` with BIRD analysis, manuscript Notes, and code tables.
- **TheBrain Scaffold**: turn the validated Excel hierarchy into empty or selectively populated Thoughts with structural Types and workflow Tags.
- **Roam JSON Export**: convert canonical BIRD objects or flat `BIRD分析` rows into a Roam-importable page/block array.
- **Monochrome Double Nine-Grid**: compose one A4 landscape printable visual with the main table of contents and three Routes on the left, and the Knowledge Index on the right.
- **Audit/Migrate**: check or convert existing TheBrain or legacy Excel rows.

## Deconstruct Text

1. Identify book, part, chapter, section, existing numbering, central question, and argument order.
2. Preserve existing addresses. Use `待編` only when no reliable address exists; do not silently renumber.
3. Treat `章` and `節` as structural containers. Treat `項` as the normal manuscript unit: one claim, explanation, evidence/example, and transition.
4. Split when the claim, reader question, independent case, definition, or reusable reasoning step changes. Do not split mechanically by paragraphs.
5. Target 700-1000 Chinese characters per standard item Note; allow 300-600 for bridges/definitions and up to 1200 for an indivisible case.
6. Build one BIRD object per item:
   - assign one `B`;
   - assign one `I.W`, one `I.T`, one canonical `I.K`, and zero or more deduplicated `I.A`;
   - assign two to five meaningful `R` targets when supported;
   - preserve verified `D` byte-for-byte or leave it blank/pending for proposed nodes.
7. Preserve the author's stance, metaphors, examples, and evidence. Mark missing material `待補`; never fabricate support.
8. Return a structure tree, BIRD table, JSON objects, and item Note drafts unless the user requests only one format.

Read [`references/manuscript-structuring.md`](references/manuscript-structuring.md) for segmentation and output templates.

## Assign the Index

For every node, apply this order:

1. Choose `K`: one canonical retrieval term, 4-10 Chinese characters or 2-4 English words.
2. Choose `T`: one code from the controlled Index Type table.
3. Choose `W`: importance to this specific book, not general fame.
4. Add `A`: genuine synonyms, translations, abbreviations, or established alternate spellings. Exclude the canonical keyword itself.
5. Assign the optional BIRD 2.1 `Role` only when graph or hierarchy evidence supports it; otherwise leave blank.

Do not turn every noun into an Index. Create an index object only when the term supports retrieval, interpretation, or routing.

## Build Routes

- Use `B` for the table-of-contents location and `R` for meaningful cross-links.
- Prefer ordered routes such as `FIRE -> 語意索引 -> AI對話`.
- Store Route targets as existing Thought names, Book Addresses, or verified Deep Links.
- Exclude self-links, direct hierarchy repetitions, and nearby siblings without a semantic reason.
- Mark a proposed target `待建` rather than inventing its address.

## Preserve Deep Links

Copy each supplied `D` exactly from `brain://` through the final character. Never decode, encode, normalize, shorten, repair, rename, or regenerate its slug. For a proposed Thought, use a blank cell in Excel and `待建立 Thought 後貼入` in prose.

Before returning, compare every displayed `D` with its source string. Do not claim tool verification unless a tool returned that exact value in the current run.

## Create the Excel Workbook

Read [`references/excel-schema.md`](references/excel-schema.md). Create these sheets:

1. `BIRD分析`: one row per knowledge node with normalized BIRD fields.
2. `拆書正文`: one row per item with Note text and BIRD JSON.
3. `代碼表`: Weight, Index Type, Semantic Role, StructuralType, and Tag dictionaries.

Use a real spreadsheet library. Freeze headers, enable filters, wrap long text, set practical widths, and validate the workbook after saving. Preserve legacy columns only in a separate `舊表對照` sheet when migration is requested.

## Export Roam Research JSON

Read [`references/roam-json-import.md`](references/roam-json-import.md), then use the deterministic converter:

```bash
node scripts/bird_to_roam_json.mjs bird.json roam-import.json
```

- Accept one canonical BIRD object, an array, or an object containing `items`, `records`, or flat Excel `rows`.
- Emit an array of Roam pages using only supported page/block keys.
- Omit block `uid` by default; add UIDs only when preserving existing block references is explicitly required and collision risk has been audited.
- Stop on duplicate generated page titles. Roam merges matching titles, so silent duplicates can add content to the wrong page.
- Preserve `D_DeepLink` byte-for-byte inside the `D::` block.
- Validate the final JSON before delivery and recommend importing into a test graph first.

## Create a Monochrome Double Nine-Grid

Read [`references/black-white-double-nine-grid.md`](references/black-white-double-nine-grid.md) whenever the user asks for a double nine-grid, print card, printable image, or left-directory/right-keyword visual.

- Use one A4 landscape page with two aligned 3×3 grids.
- Put five main contents, three meaningful Routes, and the core BIRD object in the left grid.
- Put eight Knowledge Index objects around the core Index in the right grid.
- Use a pure white background and black text/lines only. Do not use cell fills, gray backgrounds, gradients, shadows, or decorative color.
- Deliver editable SVG plus a 3508×2480 PNG. Also deliver a verified one-page A4 landscape PDF when print output is requested.
- Keep every displayed BIRD field faithful to the canonical objects. Do not invent Book Addresses or Deep Links to fill the layout.

## Audit Rules

Report the smallest correction for:

- duplicate or missing Book Address;
- invalid Weight or Index Type code;
- empty canonical Keyword;
- Alias duplicating Keyword or another Alias;
- unsupported Semantic Role;
- Route self-link or hierarchy-only repetition;
- missing or transformed Deep Link;
- `StructuralType` and address-depth conflict;
- multiple primary workflow Tags.

Do not move, rename, merge, or delete live Thoughts without explicit authorization.

## Standard BIRD Output

```text
【BIRD 2.1】
B: 全系統/第三部/3.4/3.4.D/3.4.D.B
I:
  W: I3
  T: S | Skill
  K: 八領域週檢視
  A: Weekly Review to 8 Rocks | 八岩週檢視
R: 週檢視 -> 八領域週檢視 -> 週計劃
D: brain://... or 待建立 Thought 後貼入
StructuralType: 項
Tag: 草稿
Role:
```

For Agent interchange, also output:

```json
{
  "version": "BIRD-2.1",
  "bookAddress": "全系統/第三部/3.4/3.4.D/3.4.D.B",
  "index": {
    "weight": "I3",
    "typeCode": "S",
    "type": "Skill",
    "keyword": "八領域週檢視",
    "aliases": ["Weekly Review to 8 Rocks", "八岩週檢視"]
  },
  "route": ["週檢視", "週計劃"],
  "deepLink": "brain://...",
  "structuralType": "項",
  "tag": "草稿",
  "semanticRole": null
}
```

## TheBrain Instructions

- For one current Thought, return [`references/thebrain-instruction.md`](references/thebrain-instruction.md).
- For chapter/section text deconstruction, return [`references/thebrain-manuscript-instruction.md`](references/thebrain-manuscript-instruction.md).
- For bulk Excel-to-TheBrain scaffolds, read [`references/thebrain-scaffold-import.md`](references/thebrain-scaffold-import.md) before creating any Thought, Type, Tag, or Link.