diff --git a/main.py b/main.py
new file mode 100644
index 0000000..0e58933
--- /dev/null
+++ b/main.py
@@ -0,0 +1,93 @@
+"""Task parser for raw assignment text.
+
+This module demonstrates how to convert a free‑form assignment description into a
+structured data object using LangChain and Pydantic.
+
+The main entry point is :func:`parse_task` which accepts a string and returns a
+validated :class:`TaskCard` instance.
+
+Example usage:
+
+>>> from main import parse_task
+>>> text = "Сдайте к пятнице мини-отчёт по LangChain: 2 страницы, упор на агентов. Оценка: за полноту и за пример кода."
+>>> card = parse_task(text)
+>>> print(card.model_dump())
+{"title": "мини-отчёт по LangChain", "subject": "LangChain", "deadline_hint": "к пятнице", "deliverable_type": "отчёт", "grading_hints": ["полнота", "пример кода"]}
+"""
+
+from __future__ import annotations
+
+from typing import List
+
+from pydantic import BaseModel, Field
+from langchain_core.prompts import PromptTemplate
+from langchain_openai import ChatOpenAI
+from langchain_core.output_parsers import PydanticOutputParser
+
+
+class TaskCard(BaseModel):
+    """Structured representation of an assignment description."""
+
+    title: str = Field(..., description="Short title of the assignment")
+    subject: str = Field(..., description="Subject or topic of the assignment")
+    deadline_hint: str = Field(..., description="Free‑form deadline hint")
+    deliverable_type: str = Field(..., description="What is expected to be submitted")
+    grading_hints: List[str] = Field(..., description="List of grading criteria mentioned in the text")
+
+
+# Prompt template – we ask the model to output JSON that matches TaskCard.
+prompt_template = (
+    "You are an assistant that extracts structured information from a short assignment description."
+    " Return a JSON object with the following fields: title, subject, deadline_hint, deliverable_type, grading_hints."
+    " Do not add any extra keys or text."
+    " Example input: {input_text}\n"
+    " {format_instructions}"
+)
+
+parser = PydanticOutputParser(pydantic_object=TaskCard)
+prompt = PromptTemplate(
+    template=prompt_template,
+    input_variables=["input_text"],
+    partial_variables={"format_instructions": parser.get_format_instructions()},
+)
+
+# Use the default OpenAI model; the user can set OPENAI_API_KEY.
+llm = ChatOpenAI(temperature=0)
+
+# Chain: prompt -> LLM -> parser
+chain = prompt | llm | parser
+
+
+def parse_task(text: str) -> TaskCard:
+    """Parse raw assignment text into a :class:`TaskCard`.
+
+    Parameters
+    ----------
+    text:
+        Raw assignment description.
+
+    Returns
+    -------
+    TaskCard
+        Validated structured data.
+    """
+    return chain.invoke({"input_text": text})
+
+
+if __name__ == "__main__":
+    import sys
+
+    if len(sys.argv) < 2:
+        print("Usage: python main.py '<assignment text>'")
+        sys.exit(1)
+    raw = sys.argv[1]
+    card = parse_task(raw)
+    print("Parsed card:\n", card.model_dump(indent=2))
+    print("\nHuman‑readable summary:\n")
+    print(f"Title: {card.title}")
+    print(f"Subject: {card.subject}")
+    print(f"Deadline hint: {card.deadline_hint}")
+    print(f"Deliverable type: {card.deliverable_type}")
+    print("Grading hints:")
+    for hint in card.grading_hints:
+        print(f"- {hint}")