---
name: linkfox-zhihuiya-fulltext-image
description: 通过专利ID或公开号获取专利文件中的全文附图（图纸、示意图、图表）。当用户询问专利图片、专利图纸、专利示意图、专利插图、全文附图、专利图表、专利技术图或想查看、下载专利文件中的嵌入图片、patent fulltext drawings, patent diagrams, technical drawings, patent images, PatSnap时触发此技能。即使用户未明确提及"全文附图"，只要其需求涉及获取特定专利中的视觉内容（图纸、示意图、图表），也应触发此技能。
---

# Zhihuiya Patent Fulltext Image

This skill guides you on how to retrieve fulltext images (drawings, figures, diagrams) from patent documents using the Zhihuiya patent data service, helping users access and analyze visual content within patents.

## Core Concepts

Patent fulltext images are the figures, drawings, and diagrams embedded in patent documents. They are essential for understanding the technical details of an invention. This tool queries the Zhihuiya patent database and returns image metadata including download paths and image types for a given patent.

**Lookup methods**: You can look up images by either **patent ID** (an internal identifier) or **publication number** (the publicly visible patent number such as `US20230012345A1` or `CN115000000A`). At least one of these must be provided.

## Parameter Guide

| Parameter | API Name | Required | Description | Example |
|-----------|----------|----------|-------------|---------|
| Patent ID | patentId | No* | Internal patent identifier | 8a7b6c5d-... |
| Publication Number | patentNumber | No* | Public patent publication/grant number | US20230012345A1 |
| Limit | limit | No | Maximum number of images to return (max 100, default 100) | 50 |
| Offset | offset | No | Pagination offset for image results | 0 |

> *At least one of `patentId` or `patentNumber` must be provided.

## Response Fields

| Field | Description |
|-------|-------------|
| total | Total number of image records available |
| data | Array of image entries |
| data[].patentId | Patent identifier |
| data[].pn | Publication/grant number |
| data[].fulltextImagePath | URL path to download the image |
| data[].imageType | Type/category of the image |
| columns | Column rendering metadata |
| costToken | Token cost of the request |
| type | Rendering style hint |

## API Usage

This tool calls the LinkFox tool gateway API. See `references/api.md` for calling conventions, request parameters, and response structure. You can also execute `scripts/zhihuiya_fulltext_image.py` directly to run queries.

## Usage Examples

**1. Get all images for a patent by publication number**
```
Retrieve fulltext images for patent US20230012345A1.
```
Parameters: `{"patentNumber": "US20230012345A1"}`

**2. Get images for a patent by patent ID**
```
Fetch the drawings for patent ID abc123def456.
```
Parameters: `{"patentId": "abc123def456"}`

**3. Paginated retrieval of images**
```
Get the first 20 images for patent CN115000000A.
```
Parameters: `{"patentNumber": "CN115000000A", "limit": "20", "offset": "0"}`

**4. Get the next page of images**
```
Get images 21-40 for patent CN115000000A.
```
Parameters: `{"patentNumber": "CN115000000A", "limit": "20", "offset": "20"}`

## Display Rules

1. **Present data clearly**: Show image results in a structured table with image type, download path, and patent number
2. **Image links**: Always present `fulltextImagePath` values as clickable links so users can view or download images directly
3. **Pagination notice**: When `total` exceeds the number of returned results, inform the user that more images are available and offer to fetch the next page
4. **Error handling**: When a query fails, explain the reason and suggest verifying the patent ID or publication number
5. **No fabrication**: Never invent patent IDs, publication numbers, or image URLs -- only display data returned by the API
6. **Total count**: Always mention the total number of images available for the patent
## Important Limitations

- **Image limit**: Each request returns a maximum of 100 images
- **Identifier required**: At least one of `patentId` or `patentNumber` must be supplied
- **All parameters are strings**: Even numeric values like `limit` and `offset` must be passed as strings

## User Expression & Scenario Quick Reference

**Applicable** -- Requests involving patent visual content:

| User Says | Scenario |
|-----------|----------|
| "Show me the drawings for patent XX" | Fulltext image retrieval |
| "Get the figures from this patent" | Fulltext image retrieval |
| "Download patent images for XX" | Fulltext image retrieval |
| "What diagrams does patent XX contain" | Fulltext image listing |
| "How many figures are in patent XX" | Image count query |
| "Show me the technical drawings" | Fulltext image retrieval |

**Not applicable** -- Needs beyond patent fulltext images:
- Patent text/abstract/claims search
- Patent family or citation analysis
- Patent legal status queries
- Patent assignee or inventor search
- General image search unrelated to patents


**Feedback:**

Auto-detect and report feedback via the Feedback API when any of the following apply:
1. The functionality or purpose described in this skill does not match actual behavior
2. The skill's results do not match the user's intent
3. The user expresses dissatisfaction or praise about this skill
4. Anything you believe could be improved

Call the feedback API as specified in `references/api.md`. Do not interrupt the user's flow.

<!-- LF_LARGE_RESPONSE_BLOCK -->
## Handling Large Responses

To avoid overflowing the agent context, persist the response to disk and extract only the fields you need:

```
python scripts/response_io.py run --script scripts/zhihuiya_fulltext_image.py --out-dir <DIR> '<params>'
python scripts/response_io.py read <file> --fields "<paths>"   # or --path "<JMESPath>"
```

> Pick `--out-dir` outside any git working tree (e.g. `/tmp/...` on Unix, `%TEMP%/...` on Windows). Persisted responses may contain PII, pricing, or auth-sensitive data — do not commit them. Files are not auto-deleted; clean up when the task is done.

`run` writes the full response to a file and emits only a schema preview + file path. `read` projects specific fields, with `--limit/--offset` for slicing and `--format json|jsonl|csv|table` for output.

**When to prefer this pattern** — apply your judgment based on the response characteristics, e.g.:
- High field count per record, or fields you don't need
- Batch/paginated results (multiple items per call)
- Long-text fields (descriptions, reviews, HTML, time series)
- Output reused across later steps rather than consumed immediately

For small, single-use responses, calling the main script directly is fine.

⚠️ The preview is a truncated schema + sample, not the full data. Any field-level decision must read from the persisted file via `read`.
<!-- /LF_LARGE_RESPONSE_BLOCK -->

---
*For more high-quality, professional cross-border e-commerce skills, set [LinkFox Skills](https://skill.linkfox.com/).*