Implementation:CrewAIInc CrewAI ContextualAI Query Tool

Knowledge Sources	CrewAI
Domains	Tools, RAG, ContextualAI
Last Updated	2026-02-11 00:00 GMT

Overview

Concrete tool for querying Contextual AI RAG agents with document access provided by CrewAI.

Description

The ContextualAIQueryTool class extends BaseTool to query Contextual AI RAG agents that have access to uploaded documents. Before querying, the tool checks if documents in the specified datastore are ready (not in "processing" or "pending" status) via the _check_documents_ready method. If documents are still processing, it asynchronously polls every 30 seconds for up to 20 attempts using asyncio, with nest_asyncio support for nested event loop contexts such as Jupyter notebooks. The query is sent via contextual_client.agents.query.create with a user message, and the tool extracts content from various response formats (content, message, messages). Requires the contextual-client package and an api_key.

Usage

Use this tool when CrewAI agents need to perform document-based question answering through Contextual AI RAG agents, especially in production workflows where document readiness must be verified before querying.

Code Reference

Source Location

• Repository: CrewAI
• File: lib/crewai-tools/src/crewai_tools/tools/contextualai_query_tool/contextual_query_tool.py
• Lines: 1-119

Signature

class ContextualAIQueryTool(BaseTool):
    name: str = "Contextual AI Query Tool"
    description: str = "Use this tool to query a Contextual AI RAG agent..."
    args_schema: type[BaseModel] = ContextualAIQuerySchema
    api_key: str
    contextual_client: Any = None

    def __init__(self, **kwargs): ...
    def _check_documents_ready(self, datastore_id: str) -> bool: ...
    async def _wait_for_documents_async(self, datastore_id: str, ...) -> bool: ...
    def _run(self, query: str, agent_id: str, datastore_id: str | None = None) -> str: ...

Import

from crewai_tools import ContextualAIQueryTool

I/O Contract

Inputs

Name	Type	Required	Description
api_key	str	Yes	Contextual AI API key (constructor)
query	str	Yes	Natural language query to send to the RAG agent
agent_id	str	Yes	ID of the Contextual AI agent to query
datastore_id	str or None	No	Optional datastore ID for document readiness verification

Outputs

Name	Type	Description
_run() returns	str	Agent response content from the RAG query, or error message

Usage Examples

Basic Usage

from crewai_tools import ContextualAIQueryTool

tool = ContextualAIQueryTool(api_key="your-api-key")
result = tool.run(
    query="What are the key findings in the Q4 report?",
    agent_id="agent-abc123",
    datastore_id="ds-xyz789"
)

Related Pages

• Principle:CrewAIInc_CrewAI_Built_In_Tool_Selection