zhengyanglsun/iflow-search-py

★ 14PythonAudience · developerComplexity · 2/5Setup · easy

Mindmap

mindmap
  root((iflow-search-py))
    Features
      Web search
      Image search
      Page fetching
    Usage
      Python SDK
      Async support
      MCP adapter
    AI Integration
      Claude Desktop
      Claude Code
      LangChain planned
    Security
      API key via env var
      No key in code
      100 plus test cases

mindmap root((iflow-search-py)) Features Web search Image search Page fetching Usage Python SDK Async support MCP adapter AI Integration Claude Desktop Claude Code LangChain planned Security API key via env var No key in code 100 plus test cases

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Add real-time web search to an AI agent so it can look up current information instead of relying only on its training data.

USE CASE 2

Install the MCP adapter to give Claude Desktop or Claude Code live web search and page fetching without writing any custom code.

USE CASE 3

Fetch and parse a specific URL's content in a Python script to feed live web data into a language model pipeline.

Tech stack

PythonMCPasyncioPyPI

Getting it running

Difficulty · easy Time to first run · 30min

Requires an iFlow Search API key, install with the --pre flag since the package is currently a pre-release on PyPI.

In plain English

This repository is a Python software development kit that connects AI agents and language model applications to the iFlow Search API, a service that provides web search, image search, and web page retrieval. The SDK is aimed at developers building applications that need their AI systems to look things up on the internet in real time, rather than relying only on the model's built-in knowledge. The core package lets you write a few lines of Python to search the web for a topic, find images by keyword, or fetch the content of a specific URL, all returning structured data ready for a language model to read. Both synchronous and asynchronous usage patterns are supported, so it fits into straightforward scripts as well as more complex async applications. The package is published on PyPI under the name iflow-search, though it is currently a pre-release and requires a flag when installing. Alongside the core SDK, the repository also ships an MCP adapter, published as iflow-search-mcp. MCP is a protocol that lets AI assistants like Claude Desktop or Claude Code call external tools during a conversation. Installing this adapter and pointing your AI host at it gives the assistant access to web search and page fetching without any extra coding on your part. The project follows careful security practices: the API key is read from an environment variable only and is never embedded in code, logs, or configuration files. All three search endpoints have been verified against the live API, and the test suite covers over 100 offline cases with strict type checking and linting in place. Future adapters for LangChain and OpenAPI-based platforms like Open WebUI are planned but not yet released. The repository layout is clean and well-documented, with design documents covering both the core SDK and the MCP adapter.

Copy-paste prompts

Prompt 1

I have an iFlow Search API key. Write a Python script using iflow-search that searches for today's top AI news and summarizes the top five results.

Prompt 2

Help me add the iflow-search-mcp adapter to my Claude Desktop config file so my assistant can search the web during conversations.

Prompt 3

I am building an async Python agent. Show me how to use iflow-search's async client to run a web search and an image search in parallel and combine the results into one response.

Prompt 4

I want to wrap iflow-search in a LangChain Tool. Write a custom LangChain Tool class that calls the iflow-search Python SDK for web search.

Open on GitHub → Explain another repo

← zhengyanglsun on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.