word-table-figure

name	word-table-figure
description	Table and figure module for academic paper Word documents. Creates three-line tables, formats existing tables, inserts images with captions, and manages numbering consistency. Triggers: 表格, 三线表, 图片, 插图, 题注, table, figure, image, caption.
allowed-tools	Read Bash Glob Grep mcp__word-document-server__add_table mcp__word-document-server__format_table mcp__word-document-server__format_table_cell_text mcp__word-document-server__set_table_width mcp__word-document-server__set_table_column_widths mcp__word-document-server__auto_fit_table_columns mcp__word-document-server__merge_table_cells_horizontal mcp__word-document-server__merge_table_cells_vertical mcp__word-document-server__highlight_table_header mcp__word-document-server__set_table_cell_shading mcp__word-document-server__set_table_cell_alignment mcp__word-document-server__set_table_cell_padding mcp__word-document-server__set_table_alignment_all mcp__word-document-server__apply_table_alternating_rows mcp__word-document-server__add_picture mcp__word-document-server__add_paragraph mcp__word-document-server__create_custom_style mcp__word-document-server__find_text_in_document mcp__word-document-server__get_paragraph_text_from_document mcp__word-mcp-live__insert_equation
metadata	{"version":"1.1.0","category":"editing","upstream-skills":["word-read"],"downstream-skills":["word-check"]}

Word Table & Figure

Overview

This module creates and formats tables and figures in academic paper Word documents. Its signature feature is one-command three-line table generation following academic conventions.

When to Use

• Creating a new table (especially three-line tables)
• Formatting an existing table (column widths, alignment, borders)
• Inserting an image with a formatted caption
• Merging table cells
• Adjusting table/figure numbering

When NOT to Use

• Changing document-wide formatting → use word-format
• Editing text content → use word-edit
• Checking figure/table cross-references → use word-check

---

Feature A: Three-Line Table Creation

Input

User provides:

• Table data (header row + data rows)
• Optional: column widths, alignment, caption text
• Optional: table number (or auto-detect next number from Document Map)

Process

Step 1: Add caption paragraph (style-first — use the Caption style, never run-level direct formatting)
  → add_paragraph: "表{n} {caption_text}", style="Caption"
  → Ensure the Caption style matches the Format Spec (one call covers ALL captions):
     Bash: python3 scripts/format_document.py "{file_path}" styles \
       --name Caption --cn-font 宋体 --en-font "Times New Roman" \
       --font-size 五号 --alignment center
     (或在 Caption 样式不存在时用 create_custom_style 创建)

Step 2: Create table
  → add_table: create table with header + data rows

Step 3: Apply three-line style
  → format_table: remove all borders first
  → format_table: set top border (1.0pt solid black)
  → format_table: set header-bottom border (0.5pt solid black)
  → format_table: set bottom border (1.0pt solid black)

Step 4: Format content
  → set_table_cell_alignment: center header, left/right-align data as needed
  → set_table_cell_padding: add consistent internal padding
  → format_table_cell_text: apply font per Format Spec
    （注：这是“表格单元格内文字”的特例 —— 单元格文字不走段落样式体系，
     允许使用 format_table_cell_text；这不属于被禁止的 format_text 直接格式化。
     操作完成后仍必须执行 Post-Execution 的字体归一化。）

Step 5: Set dimensions
  → set_table_width: full content width
  → set_table_column_widths: user-specified or auto-fit

Three-Line Table Border Specification

Border positions and weights:
┌─────────────────────────────────┐  ← Top line: 1.0-1.5pt solid black
│  Header 1  │  Header 2  │  ... │
├─────────────────────────────────┤  ← Header line: 0.5-0.75pt solid black
│  Data 1    │  Data 2    │  ... │
│  Data 3    │  Data 4    │  ... │
└─────────────────────────────────┘  ← Bottom line: 1.0-1.5pt solid black

No vertical lines.
No other horizontal lines between data rows.

Feature B: Format Existing Table

Input

• Table location (from Document Map or user description)
• Desired format (three-line, column widths, alignment)

Process

1. Locate table by index or nearby text
2. Apply requested formatting changes
3. Optionally convert to three-line style

Feature C: Figure Insertion

Input

• Image file path
• Caption text
• Figure number (or auto-detect)
• Optional: width/height constraints

Process

Step 1: Insert image
  → add_picture: insert at specified location with dimensions

Step 2: Add caption below (style-first — use the Caption style)
  → add_paragraph: "图{n} {caption_text}", style="Caption"
  → Caption 样式不符合 Format Spec 时，统一调整样式定义（同 Feature A Step 1）:
     Bash: python3 scripts/format_document.py "{file_path}" styles \
       --name Caption --cn-font 宋体 --en-font "Times New Roman" \
       --font-size 五号 --alignment center

Feature D: Cell Merging

Input

• Table location
• Cells to merge (row/column ranges)
• Direction (horizontal or vertical)

Process

→ merge_table_cells_horizontal or merge_table_cells_vertical
→ Re-apply alignment to merged cell

Feature E: Numbering Check

Before creating a new table/figure:

1. Read Document Map assets section
2. Determine next sequential number
3. If user specifies a number, verify it doesn't conflict

---

Post-Execution

After creating or modifying tables/figures:

1. MANDATORY: Font Normalization — Run python3 scripts/normalize_fonts.py "{file_path}" --unify before returning the document. See ../../references/font_normalization.md.
2. Report what was done
3. Suggest running word-checker to verify cross-references still valid

Shared Resources

• ../../references/academic_formatting.md — Three-line table specification, caption conventions
• ../../references/tool_routing.md — Tool selection priority