linkedapi 1.2.0

Official synchronous Python SDK for Linked API.

This official Python SDK is the synchronous way to integrate with Linked API. It provides pre-defined workflow operations, direct account/statistics methods, admin APIs, authentication headers, polling, and structured error responses.

Get started

To get started with Linked API Python SDK, read these essential guides first:

1. Core concepts - understand how Linked API works.
2. Installation & authorization - install the SDK and authorize it.
3. Predefined vs. custom workflows - choose ready-made operations or custom workflow definitions.
4. Handling results and errors - process successful data and action errors.
5. Persisting and cancelling workflows - save, resume, and cancel workflows.

Install the package:

pip install linkedapi

Initialize the client:

from linkedapi import LinkedApi, LinkedApiConfig

linkedapi = LinkedApi(
    LinkedApiConfig(
        linked_api_token="your-linked-api-token",
        identification_token="your-identification-token",
    )
)

Run a predefined workflow and poll the result:

from linkedapi import FetchPersonParams

workflow = linkedapi.fetch_person.execute(
    FetchPersonParams(
        person_url="https://www.linkedin.com/in/john-doe",
        retrieve_experience=True,
        retrieve_posts=True,
        posts_retrieval_config={"limit": 10},
    )
)

result = linkedapi.fetch_person.result(workflow.workflow_id)
if result.data:
    print(result.data.name)
for error in result.errors:
    print(error.type, error.message)

Execute a custom workflow:

workflow = linkedapi.custom_workflow.execute(
    {
        "actionType": "st.searchCompanies",
        "term": "Tech Inc",
        "filter": {
            "sizes": ["51-200", "2001-5000"],
            "locations": ["San Francisco", "New York"],
            "industries": ["Software Development"],
        },
        "then": {"actionType": "st.openCompanyPage", "basicInfo": True},
    }
)

result = linkedapi.custom_workflow.result(workflow.workflow_id)
print(result.data)

Continue polling after a workflow timeout:

from linkedapi import LinkedApiWorkflowTimeoutError

try:
    result = linkedapi.fetch_person.result(workflow.workflow_id, timeout=30.0)
except LinkedApiWorkflowTimeoutError as error:
    result = linkedapi.fetch_person.result(error.workflow_id)

Cancel a workflow:

cancelled = linkedapi.fetch_person.cancel(workflow.workflow_id)
print(cancelled)

License

This project is licensed under the MIT License. See LICENSE for details.