iflow-search-crewai
CrewAI tools for iFlow Search (心流搜索) — web search, image search, and web-page fetching, exposed as native crewai.tools.BaseTool instances.
This package is a CrewAI-native Python tool adapter. It is not a CrewAI Studio / AMP official provider tile. To appear as a first-class “iFlow Search” integration inside CrewAI’s product UI, a separate official partnership / Catalog / Marketplace route is required.
- Core SDK:
iflow-search - Sibling adapters:
iflow-search-mcp,iflow-search-langchain - API docs: https://platform.iflow.cn/docs/
Install
pip install iflow-search-crewai
Configuration
export IFLOW_API_KEY="YOUR_IFLOW_API_KEY"
Optional:
export IFLOW_BASE_URL="https://platform.iflow.cn"
export IFLOW_TIMEOUT_MS="30000"
Never commit API keys. Do not hard-code keys in source files.
Basic CrewAI usage
from crewai import Agent
from iflow_search_crewai import create_iflow_search_tools
agent = Agent(
role="Researcher",
goal="Research current information from the web",
backstory="You are a careful research assistant.",
tools=create_iflow_search_tools(),
)
Single-tool usage
from iflow_search_crewai import IFlowWebSearchTool
web_search = IFlowWebSearchTool()
result = web_search.run(query="latest AI agent frameworks", count=5)
Tools
| Class | Tool name | Purpose |
|---|---|---|
IFlowWebSearchTool |
iflow_web_search |
Web search by query |
IFlowImageSearchTool |
iflow_image_search |
Image search by query |
IFlowWebFetchTool |
iflow_web_fetch |
Fetch readable content from a URL |
Arguments
| Tool | Parameters |
|---|---|
| Web Search | query: str, count: int = 10 (1–50) |
| Image Search | query: str, count: int = 10 (1–50) |
| Web Fetch | url: str (http:// or https://) |
Each tool returns a JSON string with compact, LLM-friendly fields (titles, URLs, snippets, etc.).
DeepSeek + CrewAI smoke example
Requires crewai, this package, and API keys in the environment. Do not hard-code keys.
import os
from crewai import Agent, Crew, Task, LLM
from iflow_search_crewai import IFlowWebSearchTool
llm = LLM(
model="deepseek/deepseek-chat",
api_key=os.environ["DEEPSEEK_API_KEY"],
)
researcher = Agent(
role="Web Researcher",
goal="Answer questions using live web search",
backstory="You use iflow_web_search when current information is needed.",
tools=[IFlowWebSearchTool()],
llm=llm,
verbose=True,
)
task = Task(
description="What is CrewAI? Give a one-sentence answer grounded in search results.",
expected_output="A one-sentence summary with a source URL.",
agent=researcher,
)
crew = Crew(agents=[researcher], tasks=[task], verbose=True)
result = crew.kickoff()
print(result)
If your CrewAI version prefers a different DeepSeek model string, adjust model= to match your installed CrewAI / LiteLLM configuration.
vs MCP route
iflow-search-crewai |
iflow-search-mcp / @iflow-ai/search-mcp |
|
|---|---|---|
| Integration style | Native CrewAI BaseTool |
MCP stdio server |
| Host setup | pip install + env var |
npx or Python MCP binary + MCP config |
| Best for | Python-first CrewAI projects | MCP-native hosts and MCPServerAdapter |
Both call the same iFlow Search API via the shared iflow-search core SDK (Python) or MCP adapter (JS/Python).
Boundaries
- Not a built-in CrewAI search provider.
- Not an AMP / Studio UI tile without CrewAI official integration work.
- Does not replace
iflow-search-mcpfor MCP-based workflows.
Local development
cd packages/iflow-search-crewai
python -m pip install -e "../iflow-search"
python -m pip install -e ".[dev]"
python -m pytest -q
python -m ruff check .
python -m mypy src/iflow_search_crewai
Optional real-API smoke
export IFLOW_API_KEY="YOUR_IFLOW_API_KEY"
export IFLOW_CREWAI_SMOKE=1
python scripts/smoke_real_api.py
Metadata
Release files for iflow-search-crewai 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| iflow_search_crewai-0.1.0.tar.gz | 8.0 kB | Without provenance | Details for iflow_search_crewai-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| iflow_search_crewai-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for iflow_search_crewai-0.1.0-py3-none-any.whl |
Total release size: 19.0 kB
Release files / iflow_search_crewai-0.1.0.tar.gz
| Download URL | iflow_search_crewai-0.1.0.tar.gz |
|---|---|
| Size | 8.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
64413b0d46428da03cd6e15fe9d7ed47e74e9a915732b3cd62f93fe5033e0712
|
|
BLAKE2b-256 checksum How to use checksums |
3e7e26a29d874cb1516da88aababd1569040d66b5de5a062d6db49e458ba1423
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.13
|
Release files / iflow_search_crewai-0.1.0-py3-none-any.whl
| Download URL | iflow_search_crewai-0.1.0-py3-none-any.whl |
|---|---|
| Size | 11.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
3885a7bf20a7eaa184451d75f748d8dacd74516d729f1f2e361723029cdb1600
|
|
BLAKE2b-256 checksum How to use checksums |
e15c9bdc790580d92fc210d0fb24ca192ee3fd35c2004da49a8d3e6daa6054bb
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.13
|
Security for iflow-search-crewai 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.