Verosight Monitor — Social Media Intelligence Skill
Integrate Verosight API for real-time social media monitoring and digital reputation management across major platforms.
Quick Setup
1. Get API Key
Sign up at
verosight.com and generate an API key from the
API Keys dashboard. Keys start with
vlt_live_ (production) or
vlt_test_ (test).
2. Authenticate
Exchange your API key for a JWT token (valid 24 hours):
CODEBLOCK0
Or use the bundled auth script:
CODEBLOCK1
3. Query Data
CODEBLOCK2
API Endpoints
Posts
GET /v1/posts
Search posts across platforms with filters.
| Parameter | Type | Description |
|---|
| INLINECODE2 | string | Search keyword |
| INLINECODE3 |
string | Comma-separated:
x,instagram,tiktok,youtube,threads,news_portal |
|
limit | int | Results per page (default: 10, max: 50) |
|
days | int | Lookback period (1-90) |
|
sentiment | string | Filter:
positive,
negative,
neutral |
CODEBLOCK4
Get a single post by ID.
CODEBLOCK5
Get comments for a post.
Analytics
GET /v1/analytics/sentiment
Sentiment analysis for a keyword. Returns positive/negative/neutral breakdown, weighted percentages, and sample posts.
| Parameter | Type | Description |
|---|
| INLINECODE11 | string | Search keyword |
| INLINECODE12 |
string | Comma-separated platforms |
|
days | int | Lookback period |
|
limit | int | Number of sample posts to return |
CODEBLOCK7
Post/comment volume over time (daily breakdown).
CODEBLOCK8
Trending posts and profiles for a keyword.
Profiles
GET /v1/profiles?query=ACCOUNT&source=instagram
Search for social media profiles.
CODEBLOCK10
Get profile details (auto-crawls if missing).
CODEBLOCK11
Engagement stats for a profile.
Search
POST /v1/search
Semantic search across all content.
Body:
CODEBLOCK13
Account
GET /v1/account/balance
Check credit balance (0 cost).
CODEBLOCK15
Usage history and statistics (0 cost).
Response Format
All responses follow a standard envelope:
CODEBLOCK16
Workflows
Sentiment Analysis Report
- 1. Authenticate → get JWT
- Query
/v1/analytics/sentiment → get positive/negative/neutral breakdown - Query
/v1/posts with sentiment=negative → identify vocal accounts - Query
/v1/analytics/volume → trend over time - Compile report: overview → trend → narratives → insights → recommendations
For detailed workflow, see references/sentiment-workflow.md.
Influencer Identification
- 1. Query posts about target topic with high engagement
- Sort by likes, shares, views
- Cross-reference with sentiment (positive vs negative influencers)
- Profile top accounts for follower count and verification status
Bot Detection
- 1. Query posts about target topic across platforms
- Look for patterns:
- Multiple posts from same account in short timeframe
- Identical or near-identical content across accounts
- Coordinated posting timing
- New accounts with abnormally high activity
- 3. Flag suspicious accounts for manual review
PDF Report Generation
Generate professional reports with tables, charts, and formatted layouts using pdfkit (Node.js).
See references/pdf-template.md for complete guide.
Supported Platforms
| Platform | Coverage |
|---|
| X (Twitter) | Posts, replies, engagement metrics |
| Instagram |
Posts, captions, engagement |
| TikTok | Videos, descriptions |
| YouTube | Videos, comments |
| Threads | Posts, replies |
| Facebook | Posts, pages |
| LinkedIn | Posts, articles |
| News Portals | Articles from Indonesian media |
Credit System
Each successful API call (2xx response) costs credits. Failed calls (4xx/5xx) do not deduct credits.
| Tier | Rate Limit | Starting Credits |
|---|
| Standard | 60 req/min | 1,000 |
| Pro |
300 req/min | 10,000 |
| Enterprise | Custom | Custom |
Check balance:
CODEBLOCK17
Known Limitations
- - News portal results dominate volume (~70%); social platforms contribute less
- Semantic search returns broadly relevant results; source/date filters are limited
- Instagram direct scraping often blocked; use Verosight API for Instagram data
- Profile metadata (followers, account age) may not be available for all accounts
Bundled Resources
| File | Purpose |
|---|
| INLINECODE19 | Step-by-step sentiment analysis workflow with report template |
| INLINECODE20 |
PDF generation guide with pdfkit (tables, charts, layout) |
|
scripts/verosight-auth.sh | Auth helper — exchange API key for JWT |
|
scripts/quick-sentiment.sh | Quick sentiment check from command line |
Links
技能名称: verosight-monitor
详细描述:
Verosight Monitor — 社交媒体情报技能
集成 Verosight API,实现跨主流平台的实时社交媒体监控与数字声誉管理。
快速设置
1. 获取 API 密钥
在
verosight.com 注册,并从
API 密钥管理面板 生成 API 密钥。密钥以 vlt
live(生产环境)或 vlt
test(测试环境)开头。
2. 身份验证
将您的 API 密钥兑换为 JWT 令牌(有效期为 24 小时):
bash
JWT=$(curl -s -X POST https://api.verosight.com/v1/auth/token \
-H X-API-Key: vltliveYOUR_KEY | jq -r .token)
或使用附带的身份验证脚本:
bash
./scripts/verosight-auth.sh vltliveYOUR_KEY
3. 查询数据
bash
情感分析
curl -s https://api.verosight.com/v1/analytics/sentiment?query=KEYWORD&sources=x,instagram&days=7 \
-H Authorization: Bearer $JWT
搜索帖子
curl -s https://api.verosight.com/v1/posts?query=KEYWORD&sources=x,instagram&limit=10 \
-H Authorization: Bearer $JWT
API 端点
帖子
GET /v1/posts
跨平台搜索帖子,支持筛选。
| 参数 | 类型 | 描述 |
|---|
| query | 字符串 | 搜索关键词 |
| sources |
字符串 | 逗号分隔:x,instagram,tiktok,youtube,threads,news_portal |
| limit | 整数 | 每页结果数(默认:10,最大:50) |
| days | 整数 | 回溯天数(1-90) |
| sentiment | 字符串 | 筛选:positive、negative、neutral |
GET /v1/posts/:id
根据 ID 获取单条帖子。
GET /v1/posts/:id/comments
获取帖子的评论。
分析
GET /v1/analytics/sentiment
关键词的情感分析。返回正面/负面/中性分布、加权百分比和示例帖子。
| 参数 | 类型 | 描述 |
|---|
| query | 字符串 | 搜索关键词 |
| sources |
字符串 | 逗号分隔的平台 |
| days | 整数 | 回溯天数 |
| limit | 整数 | 返回的示例帖子数量 |
GET /v1/analytics/volume
帖子/评论随时间变化的数量(按天细分)。
GET /v1/analytics/trending
关键词的热门帖子和个人资料。
个人资料
GET /v1/profiles?query=ACCOUNT&source=instagram
搜索社交媒体个人资料。
GET /v1/profiles/:source/:account
获取个人资料详情(缺失时自动抓取)。
GET /v1/profiles/:source/:account/stats
个人资料的互动统计。
搜索
POST /v1/search
对所有内容进行语义搜索。
请求体:
json
{
query: keyword,
sources: [x, instagram],
days: 7
}
账户
GET /v1/account/balance
查询信用额度余额(免费)。
GET /v1/account/usage
使用记录和统计(免费)。
响应格式
所有响应均遵循标准信封结构:
json
{
data: { ... },
meta: {
request_id: uuid,
credits_used: 2,
credits_remaining: 998
},
pagination: {
next_cursor: base64...,
has_more: true,
total: 38113
}
}
工作流程
情感分析报告
- 1. 身份验证 → 获取 JWT
- 查询 /v1/analytics/sentiment → 获取正面/负面/中性分布
- 使用 sentiment=negative 查询 /v1/posts → 识别活跃负面账户
- 查询 /v1/analytics/volume → 随时间变化的趋势
- 编写报告:概述 → 趋势 → 叙事 → 洞察 → 建议
详细工作流程请参见 references/sentiment-workflow.md。
网红识别
- 1. 查询目标话题的高互动帖子
- 按点赞、分享、浏览量排序
- 与情感分析交叉对比(正面 vs 负面网红)
- 分析头部账户的粉丝数和认证状态
机器人检测
- 1. 跨平台查询目标话题的帖子
- 寻找模式:
- 同一账户在短时间内发布多条帖子
- 不同账户发布相同或几乎相同的内容
- 协调一致的发布时间
- 新账户异常活跃
- 3. 标记可疑账户进行人工审核
PDF 报告生成
使用 pdfkit(Node.js)生成包含表格、图表和格式化布局的专业报告。
完整指南请参见 references/pdf-template.md。
支持的平台
| 平台 | 覆盖范围 |
|---|
| X(Twitter) | 帖子、回复、互动指标 |
| Instagram |
帖子、描述、互动 |
| TikTok | 视频、描述 |
| YouTube | 视频、评论 |
| Threads | 帖子、回复 |
| Facebook | 帖子、页面 |
| LinkedIn | 帖子、文章 |
| 新闻门户 | 印尼媒体文章 |
信用额度系统
每次成功的 API 调用(2xx 响应)消耗信用额度。失败的调用(4xx/5xx)不扣除信用额度。
| 层级 | 速率限制 | 初始信用额度 |
|---|
| 标准版 | 60 次/分钟 | 1,000 |
| 专业版 |
300 次/分钟 | 10,000 |
| 企业版 | 自定义 | 自定义 |
查询余额:
bash
curl -s https://api.verosight.com/v1/account/balance \
-H Authorization: Bearer $JWT
已知限制
- - 新闻门户结果占主导地位(约 70%);社交平台贡献较少
- 语义搜索返回广泛相关的结果;来源/日期筛选有限
- Instagram 直接抓取常被阻止;请使用 Verosight API 获取 Instagram 数据
- 部分账户的个人资料元数据(粉丝数、账户年龄)可能不可用
附带资源
| 文件 | 用途 |
|---|
| references/sentiment-workflow.md | 带报告模板的分步情感分析工作流程 |
| references/pdf-template.md |
使用 pdfkit 生成 PDF 的指南(表格、图表、布局) |
| scripts/verosight-auth.sh | 身份验证辅助脚本 — 将 API 密钥兑换为 JWT |
| scripts/quick-sentiment.sh | 命令行快速情感检查 |
链接