Implementation:CrewAIInc CrewAI Linkup Search Tool

Knowledge Sources	CrewAI
Domains	Tools, Search, API_Integration
Last Updated	2026-02-11 00:00 GMT

Overview

LinkupSearchTool integrates the Linkup API search platform into CrewAI agents for retrieving contextual information from the web.

Description

The LinkupSearchTool extends BaseTool to wrap the Linkup SDK client. It provides CrewAI agents with access to Linkup's contextual search capabilities, enabling retrieval of up-to-date information during task execution. The tool handles optional installation of the linkup-sdk package via interactive confirmation using click. It initializes a LinkupClient with an API key sourced either from the constructor parameter or the LINKUP_API_KEY environment variable. The _run method executes searches with configurable depth levels (standard or deep) and output types (searchResults, sourcedAnswer, or structured), returning structured results containing name, URL, and content for each search result. All responses include a success/failure status indicator.

Usage

Use this tool when a CrewAI agent needs to perform contextual web searches through the Linkup API, retrieving real-time information with configurable search depth and result formatting.

Code Reference

Source Location

• Repository: CrewAI
• File: lib/crewai-tools/src/crewai_tools/tools/linkup/linkup_search_tool.py
• Lines: 1-81

Signature

class LinkupSearchTool(BaseTool):
    name: str = "Linkup Search Tool"
    description: str = "Performs an API call to Linkup to retrieve contextual information."
    _client: LinkupClient = PrivateAttr()
    package_dependencies: list[str] = Field(default_factory=lambda: ["linkup-sdk"])
    env_vars: list[EnvVar] = Field(default_factory=lambda: [
        EnvVar(name="LINKUP_API_KEY", description="API key for Linkup", required=True),
    ])

    def __init__(self, api_key: str | None = None) -> None: ...

    def _run(
        self,
        query: str,
        depth: Literal["standard", "deep"] = "standard",
        output_type: Literal["searchResults", "sourcedAnswer", "structured"] = "searchResults",
    ) -> dict: ...

Import

from crewai_tools import LinkupSearchTool

I/O Contract

Inputs

Name	Type	Required	Description
api_key	str or None	No	Linkup API key; falls back to LINKUP_API_KEY environment variable
query	str	Yes	The search query to execute against the Linkup API
depth	Literal["standard", "deep"]	No	Search depth level; defaults to "standard"
output_type	Literal["searchResults", "sourcedAnswer", "structured"]	No	Desired result format; defaults to "searchResults"

Outputs

Name	Type	Description
_run() returns	dict	Dictionary with "success" boolean and either "results" list (each containing name, url, content) or "error" string

Usage Examples

Basic Usage

from crewai_tools import LinkupSearchTool

tool = LinkupSearchTool(api_key="your-linkup-api-key")
result = tool._run(query="latest AI research papers")

Deep Search with Sourced Answer

from crewai_tools import LinkupSearchTool

tool = LinkupSearchTool()  # Uses LINKUP_API_KEY env var
result = tool._run(
    query="What are the latest trends in machine learning?",
    depth="deep",
    output_type="sourcedAnswer",
)

Related Pages

• Principle:CrewAIInc_CrewAI_Built_In_Tool_Selection