figure-description

Generate patent drawing descriptions and reference numeral indexes from figure files.

Updated Apr 21, 2026
One-click install
npx skills add https://github.com/Shallow-W/llm-wiki --skill figure-description-shallow-w
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: figure-description
Source: https://github.com/Shallow-W/llm-wiki/tree/main/.claude/skills/figure-description
Command: npx skills add https://github.com/Shallow-W/llm-wiki --skill figure-description-shallow-w

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps eliminate the manual drafting of patent drawing descriptions by converting user-provided patent figures into formal, reference-numeral-based explanations aligned to the patent’s disclosure and claims.

Core Features & Use Cases

  • Automated figure discovery: Scans project directories for provided patent figure files (e.g., PNG/JPG/SVG/PDF) and flags missing diagrams needed for claim coverage.
  • Reference numeral assignment: Assigns consistent 100-series numerals per figure (FIG. 1: 100–199, FIG. 2: 200–299, etc.) and links components to claim elements.
  • Jurisdiction-ready output: Produces structured drawing descriptions (Chinese or English styles) and generates a numeral index mapping numerals to components, figures, and claim mappings.
  • Use Case: When you have a set of patent drawings but need consistent “附图说明” text, this Skill generates the full descriptions and an index so every claim element is supported by a numbered component.

Quick Start

Use the figure-description skill on the provided patent figure files for the case, and generate the drawing descriptions and reference numeral index.

Frequently Asked Questions about figure-description

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate formal patent drawing descriptions from technical diagrams?

To generate formal patent drawing descriptions from technical diagrams, you process user-provided figures to assign jurisdiction-appropriate reference numerals tied to claim elements, outputting structured text and a numeral index.

How are reference numerals assigned across multiple patent figures?

Reference numerals are assigned using a 100-series consistency rule per figure, where FIG. 1 uses 100-199, FIG. 2 uses 200-299, mapping each numbered component directly to invention disclosure and claim elements.

Can I automate claims mapping for patent drafts using existing image files?

Yes, you can automate claims mapping by scanning project directories for patent figure files like PNG or PDF, analyzing image contents, and linking diagram components to specific claim elements for full coverage.

What is the best way to ensure every claim element is supported by a numbered component in patent drawings?

The best way to ensure every claim element is supported is to generate a numeral index that maps reference numerals to components, figures, and claim mappings, flagging any missing diagrams needed for coverage.

Does automated figure description generation support both Chinese and English patent styles?

Automated figure description generation supports jurisdiction-ready output by producing structured drawing descriptions in either Chinese or English styles alongside a comprehensive numeral index mapping.

What file formats are supported for patent figure discovery and processing?

Patent figure discovery and processing support image file formats including PNG, JPG, SVG, and PDF, scanning project directories to locate provided patent figures and flag missing diagrams.