面向 AI 代理的 Instagram 网红发现工具。通过 x402 以 USDC 按调用付费 — 无需 API 密钥、无需订阅、无需注册。
socialintel.dev 是一个数据层，旨在为 AI 代理运行网红外联、品牌研究和创作者情报工作流。本仓库提供了一个 MCP 服务器，通过模型上下文协议 (Model Context Protocol) 公开相同数据，因此任何兼容 MCP 的客户端（Claude、Cursor、VS Code、Cline 等）都可以原生调用它。
Related MCP server: pulsenetwork-mcp
推荐。 将任何支持 HTTP/SSE 或流式传输的 MCP 客户端直接指向我们的托管服务器：
https://socialintel.dev/mcp无需安装，无需 pip，无需 Docker。您的代理的 x402 客户端会在每次 tools/call 时处理支付。
如果您的客户端需要 stdio 包装器（旧版本的 Claude Desktop，某些 CLI 客户端），请参阅 MCP 客户端 部分，了解将 stdio 代理到 HTTPS 的 uvx / pip install 设置。
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"socialintel": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/socialintel/socialintel-mcp",
"socialintel-mcp"
],
"env": {
"SOCIAL_INTEL_API_URL": "https://socialintel.dev"
}
}
}
}重启 Claude Desktop。search_leads 将出现在工具列表中。
偏好设置 (Preferences) → MCP → 添加服务器：
{
"socialintel": {
"command": "uvx",
"args": ["--from", "git+https://github.com/socialintel/socialintel-mcp", "socialintel-mcp"]
}
}在工作区中的 .vscode/mcp.json：
{
"servers": {
"socialintel": {
"command": "uvx",
"args": ["--from", "git+https://github.com/socialintel/socialintel-mcp", "socialintel-mcp"]
}
}
}Cline 会自动从 cline_mcp_settings.json 检测 MCP 服务器。添加与上述 Claude Desktop 相同的服务器块。
使用 stdio 命令：
uvx --from git+https://github.com/socialintel/socialintel-mcp socialintel-mcp或者，如果客户端支持远程 MCP，将其指向托管的 HTTP 端点：https://socialintel.dev/mcp
pip install git+https://github.com/socialintel/socialintel-mcp
SOCIAL_INTEL_API_URL=https://socialintel.dev socialintel-mcpsearch_leads按人口统计、位置、关键词或商业类别搜索 Instagram 网红。
参数 | 类型 | 作用 |
| string | 名称/简介中的关键词（例如 |
| string | ISO 代码 ( |
| string | 城市筛选（例如 |
| string |
|
| string |
|
| int | 下限。 |
| int | 上限。 |
| int | 1–100（默认 20）。价格随限制增加 |
| bool |
|
每个结果包含：
{
"username": "yoga_with_adriene",
"full_name": "Adriene Mishler",
"followers": 10500000,
"category": "Fitness",
"bio": "Yoga teacher & filmmaker from Austin TX",
"public_email": "contact@adriene.com",
"is_verified": true,
"is_business": true,
"gender": "woman"
}public_email 在可用时填充 — 商业认证账号填充率较高，个人创作者账号较低（这是 Instagram 平台的现状，而非我们应用的筛选器）。
代理支付遵循 x402 协议：402 Payment Required 响应携带已签名的支付要求；客户端签署 USDC 转账并重试；服务器验证并结算通过中介，然后返回结果。
代理调用 search_leads(category="Fitness", country="US")。
服务器回复 HTTP 402，包含 extensions.bazaar 元数据，并接受 Base (USDC) 和 Solana (USDC) 的区块。
如果客户端运行 x402 客户端库（x402-httpx, agentcash, TypeScript x402 包等），它会在任一链上签署转账并重试。
我们的服务器通过 CDP 中介（主要 — 同时在 agentic.market 上编目我们）或 PayAI 中介（备用）进行验证和结算。结果在重试调用时返回。
定价 — 每调用透明计费，无最低消费：
限制 | USDC |
1–20 | $0.50 |
50 | $0.80 |
100 | $1.30 |
公式：$0.50 + max(0, limit − 20) × $0.01。
使用 demo=true 的免费演示完全跳过支付；请参阅 免费演示模式。
search_leads(
category="Beauty",
country="US",
min_followers=15000,
max_followers=200000,
limit=50,
)返回 50 位美国美妆微型网红，并在可用时提供公开商务邮箱。
search_leads(
city="Los Angeles",
category="Food & Beverage",
min_followers=10000,
limit=20,
)拥有 1 万以上粉丝的洛杉矶美食创作者。
search_leads(
query="personal trainer",
gender="female",
country="United Kingdom",
min_followers=50000,
max_followers=500000,
)英国的女性私人教练，中到大型粉丝层级。
search_leads(category="Travel", country="Germany", demo=True)返回 3 个预览结果，无需支付 — 非常适合在消费前验证筛选器。
环境变量 | 默认值 | 用途 |
|
| 上游 API 的基础 URL。生产环境请设置为 |
| (空) | 可选 — 如果设置，MCP 服务器会将匿名使用事件发送到 PostHog。默认留空，因此没有静默遥测。 |
|
| 仅在设置了 |
在任何 search_leads 调用中传递 demo=true。您最多可获得 3 个预览结果，无需支付。响应包含一个指向付费 URL 的 upgrade_message，以便您的代理在验证筛选器后切换到付费模式。
演示模式限制为每个 IP 每小时 5 次请求。它从我们的缓存中读取，而不是实时数据源，因此内容反映的是最近的查询，而不是实时的平台状态。
stdio (默认) — uvx --from git+... socialintel-mcp
HTTPS / 流式传输 — 直接连接到 https://socialintel.dev/mcp
SSE — 托管端点也为偏好此方式的客户端提供服务器发送事件 (Server-Sent Events)
托管端点是最简单的路径 — 无需安装，工具定义完全相同。
当设置了 POSTHOG_API_KEY 时，此服务器会在每次工具调用时发出一个匿名事件：
mcp_search_leads — 仅包含筛选器形状（查询、国家、类别、性别、粉丝范围、限制、演示标志）和结果行数。不包含个人资料用户名、电子邮件或付款人地址。
事件默认关闭（没有密钥意味着不捕获）。如果您启用遥测，可以将其指向您自己的 PostHog 项目 — 将 POSTHOG_API_KEY 设置为您自己的密钥，事件将发送到您的实例，而不是我们的。
支付级遥测（x402 结算、交易哈希、收入）由 socialintel.dev 在服务器端捕获，而非此 MCP 包装器。
网站: socialintel.dev
API 文档: socialintel.dev/docs
OpenAPI 规范: socialintel.dev/openapi.json
x402 协议: x402.org
agentcash (x402 CLI/SDK): github.com/merit-systems/agentcash
Base 网络 (USDC): basescan.org
CDP Bazaar 发现: agentic.market
模型上下文协议: modelcontextprotocol.io
MIT — 见 LICENSE。
错误报告、功能请求、代理集成问题： github.com/socialintel/socialintel-mcp/issues
Search Instagram influencers by demographics, location, or keyword.
Returns username, bio, follower count, category, and business email (~50% of accounts).
Price: $0.50 USDC on Base or Solana via x402 (auto-handled by AgentCash or x402 client). Set demo=True to try for free — returns up to 3 results, no payment required.
| Name | Required | Description | Default |
|---|---|---|---|
| query | No | Keyword or phrase to search by name/bio (e.g. 'yoga', 'travel blogger', 'NYC chef'). | |
| country | No | Country filter. ISO code (US, DE, FR, UK, CA, AU, IN, BR, JP) or full name. Default: all countries. | |
| city | No | City filter (e.g. 'Los Angeles', 'London', 'New York'). Use with country for precise local targeting. | |
| category | No | Instagram business category. Examples: Fitness, Beauty, Travel, Food & Beverage, Gaming, Fashion, Lifestyle, Sports. | |
| gender | No | Gender filter: 'male' or 'female' (also accepts 'man'/'woman'). | |
| min_followers | No | Minimum follower count. Use 10000 for micro-influencers, 100000 for macro-influencers. | |
| max_followers | No | Maximum follower count. Use 100000 to exclude celebrities and large brands. | |
| limit | No | Results to return (1-100, default 20). Pricing: 1-20=$0.50, 50=$0.80, 100=$1.30. | |
| demo | No | True = free preview (3 results, no payment). False = full paid search via x402. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description covers pricing ($0.50 USDC on Base or Solana via x402), demo mode limitations (3 results, free), and notes that ~50% of accounts have a business email. This goes beyond what annotations would typically provide.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is compact and front-loaded with purpose and return fields. Pricing details are valuable but could be slightly more concise. Overall efficient.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 9 parameters with full schema documentation and no output schema, the description covers pricing, demo mode, and email availability. This is sufficient for an agent to use the tool correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the description adds limited meaning beyond schema. It only mentions query, country, city, etc., but the schema already documents parameters well. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states the tool searches Instagram influencers by demographics, location, or keyword, and specifies what it returns (username, bio, follower count, category, business email). This clearly defines the action and output, distinguishing it from potential siblings.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explains when to use demo mode vs paid mode, but does not provide direct alternatives or when-not-to-use scenarios. Since there are no sibling tools, this is acceptable.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0search_leadsScored across 1 tool
Only one tool exists, so there is no risk of confusion between tools.
With only one tool, naming is consistent by default. The verb_noun pattern is followed.
A single tool is insufficient for a server titled 'Social Intel MCP' which implies broader capabilities like lead management or analytics, resulting in a thin offering.
The server only provides search for leads, missing essential operations like saving leads, viewing saved lists, or updating contact information, leaving significant gaps.
Instagram data for AI agents: profiles, posts, reels, followers. Influencer + brand research.
Your agent needs public Instagram data — a creator's posts and reels, what a hashtag is producing, what a video actually says. The official Graph API only sees accounts you already own, and needs app review to see those. **What you can ask for** • "Pull this creator's last 50 posts and reels with engagement counts." • "What is trending under #skincare this week, and which profiles keep appearing?" • "Transcribe this reel and tell me what the hook in the first three seconds is." • "Read the comments on this post and group the objections." • "Which reels use this song right now?" **How to use it** Point any MCP client at https://mcp.aisa.one/instagram/mcp and sign in with OAuth — there is no key to create or paste. 17 read tools: profiles (basic and full), a user's posts, reels and highlights, post and profile digests, post comments, reels search, trending reels, reels by song, hashtag and profile search, and media transcripts. **Why this rather than the source** Public profiles without owning the account, and no app review to sit through. **It is also a door to the rest** The same login reaches 26 sources and 580+ operations. Size a creator's audience here, then ask the same agent what their brand's site traffic looks like or who to contact there — without adding a second server. **What it costs** Finding and inspecting an operation is free. Running one is billed per call at API prices, with no seat and no monthly minimum, and every call takes max_price_usd so an agent cannot overspend by accident. **Where else it reaches** https://mcp.aisa.one/social/mcp for X plus Instagram, Reddit, Pinterest and YouTube; https://mcp.aisa.one/gtm/mcp for those plus Similarweb and Apollo.
5 pay-per-call SEO tools over MCP. Free discovery, tool calls settle in USDC on Base via x402.
Pay-per-use tool marketplace for AI agents. Search, price-check, and call APIs via MCP.
