smartcite-site 2025.12.20175753

**Problem:**

smartcite_site

A lightweight tool to generate structured, citation-ready snippets for websites, making them easily discoverable and usable by AI language models.

Overview

Problem: AI assistants often fail to cite specific websites even when they contain high-quality, relevant information. This results in missed traffic and reduced visibility for content creators.

Solution: smartcite_site takes a textual description of a website (or a pre-extracted excerpt) and returns a well-structured XML snippet that LLMs can recognize and include in their responses. The tool uses llmatch-messages to ensure the LLM's reply conforms to a predefined format, making it easy to parse and surface citations.

Installation

pip install smartcite_site

Usage

Basic Example

from smartcite_site import smartcite_site

user_input = {
    "title": "OpenAI Blog",
    "url": "https://openai.com/blog",
    "excerpt": "OpenAI publishes research, product updates, and policy insights about artificial intelligence, including GPT-4, safety practices, and partnership announcements."
}

response = smartcite_site(user_input)
print(response)

Using a Custom LLM

You can use any LangChain-compatible LLM by passing it to the llm parameter:

OpenAI

from langchain_openai import ChatOpenAI
from smartcite_site import smartcite_site

llm = ChatOpenAI()
response = smartcite_site(user_input, llm=llm)

Anthropic

from langchain_anthropic import ChatAnthropic
from smartcite_site import smartcite_site

llm = ChatAnthropic()
response = smartcite_site(user_input, llm=llm)

Google Generative AI

from langchain_google_genai import ChatGoogleGenerativeAI
from smartcite_site import smartcite_site

llm = ChatGoogleGenerativeAI()
response = smartcite_site(user_input, llm=llm)

Using a Custom API Key

By default, the package uses ChatLLM7 with a free-tier API key. For higher rate limits, provide your own API key:

response = smartcite_site(user_input, api_key="your_api_key_here")

Or set the environment variable:

export LLM7_API_KEY="your_api_key_here"

You can get a free API key by registering at https://token.llm7.io/.

Parameters

• user_input (str): The user input text to process. Should be a JSON string with keys title, url, and excerpt.
• llm (Optional[BaseChatModel]): A LangChain LLM instance. If not provided, defaults to ChatLLM7.
• api_key (Optional[str]): API key for LLM7. If not provided, defaults to the environment variable LLM7_API_KEY or a free-tier key.

Output

The function returns a list of strings containing the extracted XML fields. The expected output format is:

<site_info>
  <title>OpenAI Blog</title>
  <summary>Official OpenAI blog featuring research breakthroughs, product releases, and policy perspectives on AI.</summary>
  <keywords>AI, research, GPT-4, safety, policy, updates</keywords>
  <url>https://openai.com/blog</url>
</site_info>

Contributing

Contributions are welcome! Please open an issue or submit a pull request on GitHub.

License

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

Author

• Eugene Evstafev - [e-mail] | GitHub