linkup-sdk 0.14.0

A Python Client SDK for the Linkup API

🚀 Linkup Python SDK

A Python SDK for the Linkup API, allowing easy integration with Linkup's services in any Python application. 🐍

Checkout the official API documentation and SDK documentation for additional details on how to benefit from Linkup services to the full extent. 📝

🌟 Features

• ✅ Simple and intuitive API client.
• 🔍 Supports Linkup search, fetch, research, and task workflows.
• ⚡ Support synchronous and asynchronous calls.
• 🔒 Handle authentication and request management.
• 💳 Built-in x402 payment protocol support for on-chain payments.

📦 Installation

Simply install the Linkup Python SDK as any Python package, for instance using pip:

pip install linkup-sdk

To use the x402 payment protocol (on-chain payments via EVM), install the optional x402 extras:

pip install linkup-sdk[x402]

🛠️ Usage

Setting Up Your Environment

1. 🔑 Obtain an API Key:

   Sign up on Linkup to get your API key.

2. ⚙️ Set-up the API Key:

   Option 1: Export the LINKUP_API_KEY environment variable in your shell before using the Python SDK.

   export LINKUP_API_KEY=<your-linkup-api-key>
   

   Option 2: Set the LINKUP_API_KEY environment variable directly within Python, using for instance os.environ or python-dotenv with a .env file (python-dotenv needs to be installed separately in this case), before creating the Linkup Client.

   import os
   from linkup import LinkupClient
   
   os.environ["LINKUP_API_KEY"] = "<your-linkup-api-key>"
   # or dotenv.load_dotenv()
   client = LinkupClient()
   ...
   

   Option 3: Pass the Linkup API key to the Linkup Client when creating it.

   from linkup import LinkupClient
   
   client = LinkupClient(api_key="<your-linkup-api-key>")
   ...
   

📋 Examples

📝 Search

The search function can be used to performs web searches. It supports two very different complexity modes:

• with depth="standard", the search will be straightforward and fast, suited for relatively simple queries (e.g. "What's the weather in Paris today?")
• with depth="deep", the search will use an agentic workflow, which makes it in general slower, but it will be able to solve more complex queries (e.g. "What is the company profile of LangChain accross the last few years, and how does it compare to its concurrents?")

The search function also supports three output types:

• with output_type="searchResults", the search will return a list of relevant documents
• with output_type="sourcedAnswer", the search will return a concise answer with sources
• with output_type="structured", the search will return a structured output according to a user-defined schema

from typing import Any

from linkup import LinkupClient, LinkupSourcedAnswer

client = LinkupClient()  # API key can be read from the environment variable or passed as an argument
search_response: Any = client.search(
    query="What are the 3 major events in the life of Abraham Lincoln?",
    depth="deep",  # "standard" or "deep"
    output_type="sourcedAnswer",  # "searchResults" or "sourcedAnswer" or "structured"
    structured_output_schema=None,  # must be filled if output_type is "structured"
)
assert isinstance(search_response, LinkupSourcedAnswer)
print(search_response.model_dump())

Which prints:

{
  answer="The three major events in the life of Abraham Lincoln are: 1. ...",
  sources=[
    {
      "name": "HISTORY",
      "url": "https://www.history.com/topics/us-presidents/abraham-lincoln",
      "snippet": "Abraham Lincoln - Facts & Summary - HISTORY ...",
      "favicon": "https://www.history.com/favicon.ico",
    },
    ...
  ]
}

Check the code or the official documentation for the detailed list of available parameters.

🪝 Fetch

The fetch function can be used to retrieve the content of a given web page in a cleaned up markdown format.

You can use the render_js flag to execute the JavaScript code of the page before returning the content, and ask to include_raw_html to the response if you feel like it.

from linkup import LinkupClient, LinkupFetchResponse

client = LinkupClient()  # API key can be read from the environment variable or passed as an argument
fetch_response: LinkupFetchResponse = client.fetch(
    url="https://docs.linkup.so",
    render_js=False,
    include_raw_html=True,
)
print(fetch_response.model_dump())

Which prints:

{
  markdown="Get started for free, no credit card required...",
  raw_html="<!DOCTYPE html><html lang=\"en\"><head>...</head><body>...</body></html>"
}

Check the code or the official documentation for the detailed list of available parameters.

🧠 Research

The research function creates an asynchronous research task. You can then use get_research or list_research to inspect it later.

from linkup import LinkupClient, LinkupResearchTask

client = LinkupClient()
research_task: LinkupResearchTask = client.research(
    query="What changed in the AI browser market this quarter?",
    output_type="sourcedAnswer",
)
print(research_task.id)

🗂️ Tasks

The create_tasks function lets you submit mixed search, fetch, and research jobs in a single batch, then inspect them through get_task or list_tasks.

from linkup import (
    LinkupClient,
    LinkupFetchTaskInput,
    LinkupSearchTaskInput,
)

client = LinkupClient()
tasks = client.create_tasks(
    [
        LinkupSearchTaskInput(
            query="Linkup latest product updates",
            depth="deep",
            output_type="sourcedAnswer",
        ),
        LinkupFetchTaskInput(
            url="https://docs.linkup.so",
        ),
    ]
)
print([task.id for task in tasks])

⌛ Asynchronous Calls

All the Linkup main functions come with an asynchronous counterpart, with the same behavior and the same name prefixed by async_ (e.g. async_search for search). This should be favored in production use cases to avoid blocking the main thread while waiting for the Linkup API to respond. This makes possible to call the Linkup API several times concurrently for instance.

import asyncio
from typing import Any

from linkup import LinkupClient, LinkupSourcedAnswer

async def main() -> None:
    client = LinkupClient()  # API key can be read from the environment variable or passed as an argument
    search_response: Any = await client.async_search(
        query="What are the 3 major events in the life of Abraham Lincoln?",
        depth="deep",  # "standard" or "deep"
        output_type="sourcedAnswer",  # "searchResults" or "sourcedAnswer" or "structured"
        structured_output_schema=None,  # must be filled if output_type is "structured"
    )
    assert isinstance(search_response, LinkupSourcedAnswer)
    print(search_response.model_dump())

asyncio.run(main())

Which prints:

{
  answer="The three major events in the life of Abraham Lincoln are: 1. ...",
  sources=[
    {
      "name": "HISTORY",
      "url": "https://www.history.com/topics/us-presidents/abraham-lincoln",
      "snippet": "Abraham Lincoln - Facts & Summary - HISTORY ..."
      "favicon": "https://www.history.com/favicon.ico",
    },
    ...
  ]
}

💳 x402 Payment Protocol

The SDK supports the x402 payment protocol, allowing clients to automatically handle HTTP 402 responses by signing and retrying requests with on-chain payments via EVM-compatible networks (currently Base). This can be used as an alternative to API key authentication.

Create an eth_account LocalAccount compatible with Base (Ethereum):

from eth_account import Account

account = Account.from_key("<YOUR WALLET PRIVATE KEY>")

from eth_account import Account

Account.enable_unaudited_hdwallet_features()
account = Account.from_mnemonic("<YOUR MNEMONIC PHRASE>")

Then pass it to create_x402_signer and use the Linkup client:

from linkup import LinkupClient
from linkup.x402 import create_x402_signer

signer = create_x402_signer(account)
client = LinkupClient(x402_signer=signer)

result = client.search(
    query="What is x402?",
    depth="standard",
    output_type="sourcedAnswer",
)
print(result.answer)

📚 More Examples

See the examples/ directory for more examples and documentation, for instance on how to use Linkup entrypoints using asynchronous functions to call the Linkup API several times concurrenly.