[ DELIVERY METHODS ]

One source of social data. Four ways in.

Call it over HTTPS, connect an AI agent, use the TypeScript SDK, or run it from your terminal. Same data, same fields, same sub-second responses — whichever shape your stack prefers.

{ }API ◇MCP </>SDK >_CLI

{ }

[ API ]

Data as an API. Plain HTTPS.

Every platform action is a single POST endpoint that returns normalized JSON. The browser sessions, proxies, and anti-bot handling all run on our side.

✓One POST endpoint per platform action

✓Bearer-token auth, no OAuth dance

✓Every response tagged with timing & credit cost

Read the API docs →

request

curl -X POST \

https://api.bridgly.app/linkedin/get-profile-from-url \

-H "Authorization: Bearer $BRIDGLY_KEY" \

-H "Content-Type: application/json" \

-d '{"memberIdentity":"satya-nadella"}'

response200 · 428 ms

{

"timeMs": 428,

"creditsCost": 1,

"data": {

"name": "Satya Nadella",

"headline": "Chairman & CEO, Microsoft",

"followers": 10840213

}

agent sessionvia bridgly mcp

user

Find the CEO of Microsoft on LinkedIn and tell me their follower count.

tool call

linkedin_get_profile_from_url({ memberIdentity: "satya-nadella" })

tool result

{ name: "Satya Nadella", headline: "Chairman & CEO, Microsoft", followers: 10840213 }

assistant

Satya Nadella is Chairman & CEO of Microsoft. He has 10,840,213 followers on LinkedIn.

▮

[ MCP ]

Give your AI eyes on social.

Bridgly hosts a remote Model Context Protocol server, so any MCP-aware client like Claude, Cursor, or Codex connects over HTTP and pulls social data as native tools.

✓Typed tools the model discovers on its own

✓Remote server over Streamable HTTP

✓One tool per endpoint, auto-generated

mcp.json

{
  "mcpServers": {
    "bridgly": {
      "url": "https://mcp.bridgly.app/mcp",
      "headers": { "Authorization": "Bearer bgly_YOUR_KEY" }
    }
  }
}

</>

[ SDK ]

TypeScript, fully typed.

The official SDK ships ready-made types and API clients for every endpoint, compile-time safety on every response, retries and pagination handled for you.

✓Typed inputs and responses on every call

✓Errors returned as typed results, never thrown

✓ESM & CJS builds, zero config

$ npm i @bridgly/sdk

View on npm →

</> enrich.ts

import { Bridgly } from "@bridgly/sdk";

const bridgly = new Bridgly({ apiKey: process.env.BRIDGLY_KEY });

const profile = await bridgly.linkedin.getProfileFromUrl({

memberIdentity: "satya-nadella",

});

profile.▮

◆followersnumber

◆handlestring

◆headlinestring

◆locationstring

◆namestring

◆verifiedboolean

(property) Profile.followers: number

</>

[ CLI ]

Social data from your terminal.

Install the Bridgly CLI, save your API key, and call any published endpoint with a single command. Export results into your existing scripts and workflows.

✓Every published scraping endpoint available as a command

✓Searchable endpoint list and help for every flag

✓JSON, CSV, tables, and file exports

✓Read your credit balance from the terminal

Read the CLI docs →

>_ terminal

$ npm install -g @bridgly/cli

$ bridgly auth login

$ bridgly list jobs --platform linkedin

$ bridgly help reddit get-user

$ bridgly reddit get-user --username spez

$ bridgly linkedin search-jobs --keyword "software engineer"

$ bridgly github search-repositories --language TypeScript --stars.gte 100

$ bridgly balance --format table

$ bridgly reddit list-user-posts --username spez --output ./data.csv

Pick a method. Pull your first profile.

Start with 100 free credits — the same data flows through API, MCP, SDK, and CLI.

Start with 100 credits →