Douyin Search Fetch Image Search
## Purpose Search Douyin for image posts and photo collections using a text keyword. This tool retrieves search results and image links; it does not search by an uploaded image or generate images. ## Inputs Provide a required `request_body` object. The fields inside it are optional in the schema: - `keyword`: Search text; default `"猫咪"`. - `cursor`: Integer pagination cursor; default `0` for the first request. - `search_id`: Search session identifier; default `""` for the first request. - `backtrace`: Pagination backtrace value; default `""`. - `content_type`: Explicitly use the string `"2"` for image posts. The schema defaults to `"0"` and lists `"0"`=all, `"1"`=video, `"2"`=picture, `"3"`=article, but this operation is documented for image search; support for other content types is not established. - `sort_type`: `"0"`=comprehensive, `"1"`=most likes, `"2"`=latest; default `"0"`. - `publish_time`: `"0"`=unlimited, `"1"`=last day, `"7"`=last week, `"180"`=last half year; default `"0"`. - `filter_duration`: Video-duration filter: `"0"`=unlimited, `"0-1"`=within one minute, `"1-5"`=one to five minutes, `"5-10000"`=over five minutes. Default to `"0"` for image searches. For later pages, reuse the previous response's `cursor`, `search_id` and `backtrace` when available; do not invent missing pagination values. `has_more` indicates whether more results are available (`1`=yes, `0`=no). ## Example ```json {"request_body":{"keyword":"cat","cursor":0,"search_id":"","backtrace":"","content_type":"2","sort_type":"0","publish_time":"0","filter_duration":"0"}} ``` ## Results The response is returned directly and includes a `data` payload plus response metadata such as a status code and message. Documented search-result fields include pagination information and post entries containing `aweme_info`: content ID (`aweme_id`), description (`desc`), creation time, author information, `image_post_info.images` with image URL lists and pixel dimensions, engagement statistics, visibility status and a share link. These are common fields, not a guaranteed fixed structure: the payload schema is flexible and nullable, so use the actual response to determine available fields.
Pricing
| Condition | Charge |
|---|---|
| Each call | 20.65 credits/call |
20.65 credits/call Charges are per query. A successful query is billed even if it returns no results.
Parameters
Optional parameters use the platform defaults when omitted.
| Parameter | Type | Required | Description | Default |
|---|---|---|---|---|
request_body | object | Yes | 图片内容搜索的请求体对象,包含搜索关键词 keyword、翻页字段 cursor、search_id 和 backtrace,以及排序和筛选字段 sort_type、publish_time、filter_duration、content_type。用于搜索图片帖子及多图合集。 / Request body object for image-post search, containing the search keyword (keyword), pagination fields (cursor, search_id and backtrace), and sorting and filtering fields (sort_type, publish_time, filter_duration and content_type). Used to search image posts and image collections. | — |
request_body.cursor | integer | No | 偏移游标,用于翻页,从上一次请求返回的响应中获取 / Offset cursor for pagination, obtained from the last response | 0 |
request_body.keyword | string | No | 关键词 / Keyword | 猫咪 |
request_body.backtrace | string | No | 翻页回溯标识,用于翻页,从上一次请求返回的响应中获取 / Backtrace for pagination, obtained from the last response | "" |
request_body.search_id | string | No | 搜索ID,用于翻页,从上一次请求返回的响应中获取 / Search ID for pagination, obtained from the last response | "" |
request_body.sort_type | string | No | 排序方式:0=综合排序 1=最多点赞 2=最新发布 / Sort type: 0=Comprehensive, 1=Most Likes, 2=Latest | 0 |
request_body.content_type | string | No | 内容类型:0=不限 1=视频 2=图片 3=文章 / Content type: 0=All, 1=Video, 2=Picture, 3=Article | 0 |
request_body.publish_time | string | No | 发布时间筛选:0=不限 1=最近一天 7=最近一周 180=最近半年 / Publish time filter: 0=Unlimited, 1=Last day, 7=Last week, 180=Last half year | 0 |
request_body.filter_duration | string | No | 视频时长过滤:0=不限 0-1=一分钟以内 1-5=一到五分钟 5-10000=五分钟以上 / Video duration filter: 0=Unlimited, 0-1=Within 1 minute, 1-5=1 to 5 minutes, 5-10000=More than 5 minutes | 0 |
Invocation example
Copy the request and explore its response example.
/agent-api/v1/tool/executeX-Api-Keycurl --request POST \
--url 'https://agentearth.ai/agent-api/v1/tool/execute?tool_name=ae_douyin_search_fetch_image_search' \
--header "X-Api-Key: $AGENTEARTH_API_KEY" \
--header 'Content-Type: application/json' \
--data-raw '{
"params": {
"request_body": {
"backtrace": "",
"content_type": "2",
"cursor": 0,
"filter_duration": "0",
"keyword": "毛绒挂件",
"publish_time": "180",
"search_id": "",
"sort_type": "1"
}
}
}'{
"code": 200,
"data": {
"business_data": [
{
"data": {
"aweme_info": {
"author": {},
"aweme_id": "7706004578797368506",
"create_time": 1794194006,
"desc": "原神麦当劳联动!!竟然还有毛绒!\n#原神 #瓦雷莎 #爱可菲",
"music": {}
},
"type": 1
},
"data_id": "0",
"type": 11
}
]
},
"docs": "https://api.tikhub.io/#/Douyin-Search-API/fetch_image_search_api_v1_douyin_search_fetch_image_search_post",
"message": "Request successful. This request will incur a charge.",
"message_zh": "请求成功,本次请求将被计费。",
"request_id": "4b497431-d7c2-48b0-87d2-4ff032b143d2"
}This example shows part of the response. Copied content is valid JSON.

