concept‑extractor

concept‑extractor is a lightweight Python package that turns complex, philosophical or thematic text prompts (e.g., “Dark Enlightenment”) into concise, well‑structured summaries or analyses. It leverages language models via LangChain and a pattern‑matching helper (llmatch) to guarantee that the output conforms to a user‑defined regular‑expression pattern.

Key points

• Text‑only – no multimedia handling.
• Deterministic output thanks to regex‑based validation.
• Works out‑of‑the‑box with the default ChatLLM7 model; any LangChain‑compatible LLM can be swapped in.

Installation

pip install concept_extractor

Quick start

from concept_extractor import concept_extractor

# Your prompt describing a complex concept
user_input = """
The Dark Enlightenment is a contemporary philosophical movement that...
"""

# Run the extractor (uses ChatLLM7 by default)
summary = concept_extractor(user_input)

print(summary)

summary is a list of strings that matches the pattern defined in concept_extractor.prompts.

Function signature

def concept_extractor(
    user_input: str,
    api_key: Optional[str] = None,
    llm: Optional[BaseChatModel] = None,
) -> List[str]:

The function returns a List[str] containing the extracted data that matches the internal regex pattern.

Using a custom LLM

If you prefer another provider (OpenAI, Anthropic, Google, etc.), simply pass a LangChain chat model instance:

OpenAI

from langchain_openai import ChatOpenAI
from concept_extractor import concept_extractor

llm = ChatOpenAI(model="gpt-4o-mini")
response = concept_extractor(user_input, llm=llm)
print(response)

Anthropic

from langchain_anthropic import ChatAnthropic
from concept_extractor import concept_extractor

llm = ChatAnthropic(model="claude-3-haiku-20240307")
response = concept_extractor(user_input, llm=llm)
print(response)

Google Generative AI

from langchain_google_genai import ChatGoogleGenerativeAI
from concept_extractor import concept_extractor

llm = ChatGoogleGenerativeAI(model="gemini-1.5-flash")
response = concept_extractor(user_input, llm=llm)
print(response)

API key & rate limits

• The default free tier of LLM7 (the model used when you don’t supply an llm) provides generous rate limits that satisfy most use cases of this package.
• To use higher limits or your own quota, set the environment variable LLM7_API_KEY or pass the key directly:

response = concept_extractor(user_input, api_key="your_llm7_api_key")

• Obtain a free API key by registering at: https://token.llm7.io/

Contributing & support

• Issues & bug reports: https://github.com/chigwell/concept_extractor/issues
• Pull requests are welcome – see the repository for the contribution guidelines.

Author

Eugene Evstafev – hi@euegne.plus
GitHub: chigwell

License

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