{
"success": true,
"meta": {
"request_id": "req_7b21c84a901e",
"plugin": "reddit",
"endpoint": "search",
"tier": "ULTRA",
"cached": false,
"latency_ms": 722.4,
"timestamp": "2026-10-27T06:40:00.000000+00:00"
},
"data": {
"query": "openai",
"tab": "posts",
"subreddit": null,
"sort": "relevance",
"timeFilter": "all",
"pageSize": 1,
"nextCursor": "t3_1b8k9x2",
"hasMore": true,
"items": [
{
"id": "t3_1b8k9x2",
"postId": "1b8k9x2",
"title": "Next-generation AI agents and real-time community intelligence benchmarks in 2026",
"subreddit": "technology",
"author": "user_a8f2",
"url": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"score": 2840,
"upvoteRatio": 0.94,
"numComments": 342,
"createdAt": "2026-03-15T13:50:00+00:00",
"selftext": "We evaluated real-time community discussion pipelines for LLM grounding, brand sentiment tracking, and developer trend discovery.",
"postType": "text",
"flair": "Artificial Intelligence",
"isOver18": false,
"isPinned": false,
"mediaUrls": [],
"commentsCount": 0,
"comments": [],
"searchQuery": "openai",
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
}
]
}
}
curl --location --request GET 'https://zylalabs.com/api/14531/subreddit+feed+and+community+monitor+api/37112/search+reddit+posts?query=python&sort=relevance&time_range=all&limit=10&include_nsfw=false&no_cache=false' --header 'Authorization: Bearer YOUR_API_KEY'
{
"success": true,
"meta": {
"request_id": "req_8c32d95b012f",
"plugin": "reddit",
"endpoint": "subreddit",
"tier": "ULTRA",
"cached": false,
"latency_ms": 941.2,
"timestamp": "2026-10-27T06:40:00.000000+00:00"
},
"data": {
"subreddit": "technology",
"url": "https://www.reddit.com/r/technology/",
"sort": "hot",
"timeFilter": "all",
"pageSize": 1,
"nextCursor": "t3_1b8k9x2",
"hasMore": true,
"items": [
{
"id": "t3_1b8k9x2",
"postId": "1b8k9x2",
"title": "Next-generation AI agents and real-time community intelligence benchmarks in 2026",
"subreddit": "technology",
"author": "user_a8f2",
"url": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"score": 2840,
"upvoteRatio": 0.94,
"numComments": 342,
"createdAt": "2026-03-15T13:50:00+00:00",
"selftext": "We evaluated real-time community discussion pipelines for LLM grounding, brand sentiment tracking, and developer trend discovery.",
"postType": "text",
"flair": "Artificial Intelligence",
"isOver18": false,
"isPinned": false,
"mediaUrls": [],
"commentsCount": 0,
"comments": [],
"searchQuery": null,
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
}
]
}
}
curl --location --request GET 'https://zylalabs.com/api/14531/subreddit+feed+and+community+monitor+api/37113/get+subreddit+feed+posts?subreddit=python&sort=hot&time_range=all&limit=10&include_nsfw=false&no_cache=false' --header 'Authorization: Bearer YOUR_API_KEY'
{
"success": true,
"meta": {
"request_id": "req_9d43ea6c1230",
"plugin": "reddit",
"endpoint": "post",
"tier": "ULTRA",
"cached": false,
"latency_ms": 1180.5,
"timestamp": "2026-10-27T06:40:00.000000+00:00"
},
"data": {
"id": "t3_1b8k9x2",
"postId": "1b8k9x2",
"title": "Next-generation AI agents and real-time community intelligence benchmarks in 2026",
"subreddit": "technology",
"author": "user_a8f2",
"url": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"score": 2840,
"upvoteRatio": 0.94,
"numComments": 342,
"createdAt": "2026-03-15T13:50:00+00:00",
"selftext": "We evaluated real-time community discussion pipelines for LLM grounding, brand sentiment tracking, and developer trend discovery.",
"postType": "text",
"flair": "Artificial Intelligence",
"isOver18": false,
"isPinned": false,
"mediaUrls": [],
"commentsCount": 2,
"comments": [
{
"id": "kty8m1a",
"postId": "t3_1b8k9x2",
"parentId": "t3_1b8k9x2",
"subreddit": "technology",
"author": "user_4c9b",
"body": "The real latency win comes from edge caching and avoiding cold-start browser overhead on every API call.",
"score": 412,
"createdAt": "2026-03-15T14:22:10+00:00",
"depth": 0,
"isSubmitter": false,
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/sample_thread/kty8m1a/",
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
},
{
"id": "kty9b4c",
"postId": "t3_1b8k9x2",
"parentId": "t1_kty8m1a",
"subreddit": "technology",
"author": "user_7d1e",
"body": "Agreed. For RAG pipelines and AI agents, sub-second JSON responses make a huge difference compared to headless browsers.",
"score": 189,
"createdAt": "2026-03-15T14:35:42+00:00",
"depth": 1,
"isSubmitter": false,
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/sample_thread/kty9b4c/",
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
}
],
"searchQuery": null,
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
}
}
curl --location --request GET 'https://zylalabs.com/api/14531/subreddit+feed+and+community+monitor+api/37114/get+post+detail+and+top+comments?post_id=1j9x2a4&include_comments=true&comment_limit=15&no_cache=false' --header 'Authorization: Bearer YOUR_API_KEY'
{
"success": true,
"meta": {
"request_id": "req_ae54fb7d2341",
"plugin": "reddit",
"endpoint": "comments",
"tier": "ULTRA",
"cached": false,
"latency_ms": 985.1,
"timestamp": "2026-10-27T06:40:00.000000+00:00"
},
"data": {
"postId": "1b8k9x2",
"fullnameId": "t3_1b8k9x2",
"subreddit": "technology",
"postTitle": "Next-generation AI agents and real-time community intelligence benchmarks in 2026",
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"returnedCount": 2,
"maxCommentsRequested": 20,
"comments": [
{
"id": "kty8m1a",
"postId": "t3_1b8k9x2",
"parentId": "t3_1b8k9x2",
"subreddit": "technology",
"author": "user_4c9b",
"body": "The real latency win comes from edge caching and avoiding cold-start browser overhead on every API call.",
"score": 412,
"createdAt": "2026-03-15T14:22:10+00:00",
"depth": 0,
"isSubmitter": false,
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/sample_thread/kty8m1a/",
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
},
{
"id": "kty9b4c",
"postId": "t3_1b8k9x2",
"parentId": "t1_kty8m1a",
"subreddit": "technology",
"author": "user_7d1e",
"body": "Agreed. For RAG pipelines and AI agents, sub-second JSON responses make a huge difference compared to headless browsers.",
"score": 189,
"createdAt": "2026-03-15T14:35:42+00:00",
"depth": 1,
"isSubmitter": false,
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/sample_thread/kty9b4c/",
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
}
]
}
}
curl --location --request GET 'https://zylalabs.com/api/14531/subreddit+feed+and+community+monitor+api/37115/get+post+comment+thread?post_id=1j9x2a4&sort=confidence&limit=25&no_cache=false' --header 'Authorization: Bearer YOUR_API_KEY'
{
"success": true,
"meta": {
"request_id": "req_bf650c8e3452",
"plugin": "reddit",
"endpoint": "user",
"tier": "ULTRA",
"cached": false,
"latency_ms": 1120.8,
"timestamp": "2026-10-27T06:40:00.000000+00:00"
},
"data": {
"username": "user_a8f2",
"profileUrl": "https://www.reddit.com/user/user_a8f2/",
"pageSize": 1,
"nextCursor": "t3_1b8k9x2",
"hasMore": false,
"items": [
{
"id": "t3_1b8k9x2",
"postId": "1b8k9x2",
"title": "Next-generation AI agents and real-time community intelligence benchmarks in 2026",
"subreddit": "technology",
"author": "user_a8f2",
"url": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"permalink": "https://www.reddit.com/r/technology/comments/1b8k9x2/nextgeneration_ai_agents_and_realtime/",
"score": 2840,
"upvoteRatio": 0.94,
"numComments": 342,
"createdAt": "2026-03-15T13:50:00+00:00",
"selftext": "We evaluated real-time community discussion pipelines for LLM grounding, brand sentiment tracking, and developer trend discovery.",
"postType": "text",
"flair": "Artificial Intelligence",
"isOver18": false,
"isPinned": false,
"mediaUrls": [],
"commentsCount": 0,
"comments": [],
"searchQuery": null,
"scrapedAt": "2026-10-27T06:40:00.000000+00:00"
}
]
}
}
curl --location --request GET 'https://zylalabs.com/api/14531/subreddit+feed+and+community+monitor+api/37116/get+user+submitted+posts+and+activity?username=user_a8f2&tab=posts&sort=new&limit=10&include_nsfw=false&no_cache=false' --header 'Authorization: Bearer YOUR_API_KEY'
注册后,每个开发者都会被分配一个个人 API 访问密钥,这是一个唯一的字母和数字组合,用于访问我们的 API 端点。要使用 子版块动态和社区监测 API 进行身份验证,只需在 Authorization 标头中包含您的 bearer token。
| 标头 | 描述 |
|---|---|
授权
|
必需
应为 Bearer access_key. 订阅后,请查看上方的"您的 API 访问密钥"。
|
监控按热度、新发布、热门或上升排序的实时子版块信息,以跟踪社区帖子、点赞比例、评论速度、帖子标签和外部媒体链接
每个端点从公共的Reddit社区返回结构化的JSON数据:搜索Reddit帖子和获取子版块动态帖子返回带有得分和分页游标的帖子集合,获取帖子详情返回完整的线程元数据和最顶上的评论,获取帖子评论线程返回嵌套评论树,获取用户提交的帖子返回公共用户提交内容
关键帖子字段包括 `id` `title` `subreddit` `author` `score` `upvoteRatio` `commentCount` `createdAt` `permalink` `bodyText` 和 `mediaUrls` 评论项还包括 `parentId` 和 `depth` 以重建多层对话树
参数因端点而异,包括 `query` `subreddit` `sort` `time_range` `limit` `cursor` `include_nsfw` `post_id` `include_comments` `comment_limit` `username` `tab` 和 `no_cache`
当有额外项目可用时,响应会包含一个`nextCursor`字符串和`hasNextPage`布尔值。将`nextCursor`值传递到下一个请求的`cursor`参数中以检索下一页
是的。在获取帖子详细信息的端点上,将 `include_comments` 设置为 `true` 并调整 `comment_limit` 以在一次请求中接收完整的 `post` 对象和顶级线程 `comments`
每个响应使用一个一致的JSON封装,其中包含一个`success`布尔值,一个包含`request_id`、`latency_ms`和`cached`状态的`meta`对象,以及一个包含请求的`items`、`post`或`comments`的`data`对象
常见的使用案例包括人工智能代理基础和检索增强生成 社交倾听 品牌情感分析 社区趋势追踪 市场研究和开发者生态系统监测
响应默认缓存60秒,以最大化速度并保护速率限制。通过在任何端点上传递`no_cache`并设置为`true`来强制进行新鲜的实时上游提取
获取子版块帖子接口提供特定子版块帖子的结构化数据,包括帖子标题、作者、得分、赞成比例、评论数量、媒体链接和分页详情。这使用户能够有效分析社区参与度和趋势
您可以通过指定参数来过滤帖子,比如`query`用于关键词,`subreddit`用于特定社区,`sort`用于相关性、热门、顶级或最新,以及`timeFilter`用于时间窗口,如小时、天、周或所有时间
`upvoteRatio`字段表示一篇帖子中赞成票与总票数的比例,提供了社区认可度的洞察。较高的比例表明该帖子受到好评,这对于情感分析非常有用
通过提供 `post_id` 或 `fullnameId` 使用获取帖子评论线程端点 这将返回一个结构化的评论树 包括父子关系 分数 和时间戳 允许对讨论进行详细分析
如果一个端点没有返回结果,请检查您的查询参数是否准确 确保子版块存在 关键词相关 以及排序选项适当 调整这些参数可能会获得更好的结果
此端点允许您通过检索用户的公开帖子和社区互动来分析用户的活动。您可以按新、热或热门对这些数据进行排序,从而洞察用户参与和内容趋势随时间的变化
每个API响应都由一个`success`布尔值,一个包含请求细节(如`request_id`和`latency_ms`)的`meta`对象,以及一个以一致格式存放请求项(如帖子或评论)的`data`对象构成
要接收最新的数据,请在请求中使用`no_cache`参数并设置为`true`这可以绕过缓存的响应,确保您从源头直接获取最新信息