diff --git a/.github/workflows/upgrade-python-requirements.yml b/.github/workflows/upgrade-python-requirements.yml
deleted file mode 100644
index d2aef30..0000000
--- a/.github/workflows/upgrade-python-requirements.yml
+++ /dev/null
@@ -1,29 +0,0 @@
-name: Upgrade Python Requirements
-
-on:
-  schedule:
-    - cron: "0 0 * * 1"
-  workflow_dispatch:
-    inputs:
-      branch:
-        description: Target branch against which to create requirements PR
-        required: true
-        default: master
-
-jobs:
-  call-upgrade-python-requirements-workflow:
-    uses: openedx/.github/.github/workflows/upgrade-python-requirements.yml@master
-    # Do not run on forks
-    if: github.repository_owner == 'open-craft'
-    with:
-      branch: ${{ github.event.inputs.branch || 'master' }}
-      # optional parameters below; fill in if you'd like github or email notifications
-      # user_reviewers: ""
-      # team_reviewers: ""
-      # email_address: ""
-      # send_success_notification: false
-    secrets:
-      requirements_bot_github_token: ${{ secrets.REQUIREMENTS_BOT_GITHUB_TOKEN }}
-      requirements_bot_github_email: ${{ secrets.REQUIREMENTS_BOT_GITHUB_EMAIL }}
-      edx_smtp_username: ${{ secrets.EDX_SMTP_USERNAME }}
-      edx_smtp_password: ${{ secrets.EDX_SMTP_PASSWORD }}
diff --git a/Makefile b/Makefile
new file mode 100644
index 0000000..c992e2f
--- /dev/null
+++ b/Makefile
@@ -0,0 +1,98 @@
+.PHONY: clean help compile_translations dummy_translations extract_translations detect_changed_source_translations \
+		build_dummy_translations validate_translations check_translations_up_to_date \
+		requirements selfcheck test test.python test.unit test.quality upgrade
+
+.DEFAULT_GOAL := help
+
+WORKING_DIR := ai_eval
+JS_TARGET := $(WORKING_DIR)/public/js/translations
+EXTRACT_DIR := $(WORKING_DIR)/conf/locale/en/LC_MESSAGES
+EXTRACTED_DJANGO_PARTIAL := $(EXTRACT_DIR)/django-partial.po
+EXTRACTED_DJANGOJS_PARTIAL := $(EXTRACT_DIR)/djangojs-partial.po
+EXTRACTED_DJANGO := $(EXTRACT_DIR)/django.po
+
+help: ## display this help message
+	@echo "Please use \`make <target>' where <target> is one of"
+	@perl -nle'print $& if m{^[a-zA-Z_-]+:.*?## .*$$}' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m  %-25s\033[0m %s\n", $$1, $$2}'
+
+clean: ## remove generated byte code, coverage reports, and build artifacts
+	find . -name '__pycache__' -exec rm -rf {} +
+	find . -name '*.pyc' -exec rm -f {} +
+	find . -name '*.pyo' -exec rm -f {} +
+	find . -name '*~' -exec rm -f {} +
+	rm -fr build/
+	rm -fr dist/
+	rm -fr *.egg-info
+
+## Localization targets
+
+extract_translations: ## extract strings to be translated, outputting .po files
+	cd $(WORKING_DIR) && i18n_tool extract
+	mv $(EXTRACTED_DJANGO_PARTIAL) $(EXTRACTED_DJANGO)
+	# Safely concatenate djangojs if it exists. The file will exist in this repo, but we're trying to follow a pattern
+	# between all repositories that use i18n_tool
+	if test -f $(EXTRACTED_DJANGOJS_PARTIAL); then \
+	  msgcat $(EXTRACTED_DJANGO) $(EXTRACTED_DJANGOJS_PARTIAL) -o $(EXTRACTED_DJANGO) && \
+	  rm $(EXTRACTED_DJANGOJS_PARTIAL); \
+	fi
+	sed -i'' -e 's/nplurals=INTEGER/nplurals=2/' $(EXTRACTED_DJANGO)
+	sed -i'' -e 's/plural=EXPRESSION/plural=\(n != 1\)/' $(EXTRACTED_DJANGO)
+
+compile_translations: ## compile translation files, outputting .mo files for each supported language
+	cd $(WORKING_DIR) && i18n_tool generate -v
+	python manage.py compilejsi18n --namespace DragAndDropI18N --output $(JS_TARGET)
+
+detect_changed_source_translations:
+	cd $(WORKING_DIR) && i18n_tool changed
+
+dummy_translations: ## generate dummy translation (.po) files
+	cd $(WORKING_DIR) && i18n_tool dummy
+
+build_dummy_translations: dummy_translations compile_translations ## generate and compile dummy translation files
+
+validate_translations: build_dummy_translations detect_changed_source_translations ## validate translations
+
+check_translations_up_to_date: extract_translations compile_translations dummy_translations detect_changed_source_translations ## extract, compile, and check if translation files are up-to-date
+
+piptools: ## install pinned version of pip-compile and pip-sync
+	pip install -r requirements/pip.txt
+	pip install -r requirements/pip-tools.txt
+
+requirements: piptools  ## install test requirements locally
+	pip-sync requirements/ci.txt
+
+requirements_python: piptools  ## install all requirements locally
+	pip-sync requirements/dev.txt requirements/private.*
+
+test.quality: selfcheck ## run quality checkers on the codebase
+	tox -e quality
+
+test.python: ## run python unit tests in the local virtualenv
+	pytest --cov ai_eval $(TEST)
+
+test.unit: ## run all unit tests
+	tox $(TEST)
+
+test: test.unit test.quality ## Run all tests
+	tox -e translations
+
+# Define PIP_COMPILE_OPTS=-v to get more information during make upgrade.
+PIP_COMPILE = pip-compile --upgrade $(PIP_COMPILE_OPTS)
+
+upgrade: export CUSTOM_COMPILE_COMMAND=make upgrade
+upgrade: ## update the requirements/*.txt files with the latest packages satisfying requirements/*.in
+	pip install -qr requirements/pip-tools.txt
+	# Make sure to compile files after any other files they include!
+	$(PIP_COMPILE) --allow-unsafe -o requirements/pip.txt requirements/pip.in
+	$(PIP_COMPILE) -o requirements/pip-tools.txt requirements/pip-tools.in
+	pip install -qr requirements/pip.txt
+	pip install -qr requirements/pip-tools.txt
+	$(PIP_COMPILE) -o requirements/base.txt requirements/base.in
+	$(PIP_COMPILE) -o requirements/test.txt requirements/test.in
+	$(PIP_COMPILE) -o requirements/quality.txt requirements/quality.in
+	$(PIP_COMPILE) -o requirements/ci.txt requirements/ci.in
+	$(PIP_COMPILE) -o requirements/dev.txt requirements/dev.in
+	sed -i '/^[dD]jango==/d' requirements/test.txt
+
+selfcheck: ## check that the Makefile is well-formed
+	@echo "The Makefile is well-formed."
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..3c897f0
--- /dev/null
+++ b/README.md
@@ -0,0 +1,38 @@
+## Introduction
+
+This repository hosts two Open edX XBlocks: 
+
+1. **Short Answer with AI Evaluation**: This XBlock allows students to submit short answers, which are then evaluated with the help of a large language model (LLM).
+2. **Coding with AI Evaluation**: This XBlock allows students to submit code in a text editor. The code is executed via a third-party API (currently using [Judge0](https://judge0.com/)), and both the code and its output are sent to an LLM for feedback.
+
+## Screeshots
+
+| ![Short Answer with AI evaluation Xblock](docs/shortanswer-xblock.png) | ![Coding with AI evaluation Xblock](docs/coding-xblock.png) |
+|-----------------------------------------------------------------------|----------------------------------------------------------------|
+| ![Coding with AI evaluation Xblock HTML](docs/coding-xblock-ai-feedback.png) | ![Coding with AI evaluation Xblock AI feedback](docs/coding-xblock-html.png) |
+
+
+
+## Setup
+
+### Using Tutor
+
+1. Add the following line to the `OPENEDX_EXTRA_PIP_REQUIREMENTS` in your Tutor `config.yml` file:
+   ```yaml
+   OPENEDX_EXTRA_PIP_REQUIREMENTS:
+     - git+https://github.com/open-craft/xblock-ai-evaluation
+   ```
+   You can append  `@vX.Y.Z` to the URL to specify your desired version.
+
+2. Launch Tutor.
+
+3. In the Open edX platform, navigate to `Settings > Advanced Settings` and add `shortanswer_ai_eval` and `coding_ai_eval` to the `Advanced Module List`.
+
+4. Add either XBlock using the `Advanced` button in the `Add New Component` section of Studio.
+
+5. Configure the added Xblock and make sure to add correct API keys. You can format your question and prompts using [Markdown](https://marked.js.org/demo/).
+
+## Dependencies
+- [Judge0 API](https://judge0.com/)
+- [Monaco editor](https://github.com/microsoft/monaco-editor)
+- [LiteLLM](https://github.com/BerriAI/litellm)
\ No newline at end of file
diff --git a/ai_eval/__init__.py b/ai_eval/__init__.py
new file mode 100644
index 0000000..63412c2
--- /dev/null
+++ b/ai_eval/__init__.py
@@ -0,0 +1,6 @@
+"""
+Xblock to have short text and code entries with AI-driven evaluation.
+"""
+
+from .shortanswer import ShortAnswerAIEvalXBlock
+from .coding_ai_eval import CodingAIEvalXBlock
diff --git a/ai_eval/base.py b/ai_eval/base.py
new file mode 100644
index 0000000..3e67e7f
--- /dev/null
+++ b/ai_eval/base.py
@@ -0,0 +1,146 @@
+"""Base Xblock with AI evaluation."""
+
+import pkg_resources
+
+from django.utils.translation import gettext_noop as _
+from xblock.core import XBlock
+from xblock.fields import String, Scope, Dict
+from xblock.validation import ValidationMessage
+
+
+from .llm import SupportedModels
+
+try:
+    from xblock.utils.studio_editable import StudioEditableXBlockMixin
+except ModuleNotFoundError:  # For compatibility with Palm and earlier
+    from xblockutils.studio_editable import StudioEditableXBlockMixin
+
+try:
+    from xblock.utils.resources import ResourceLoader
+except (
+    ModuleNotFoundError
+):  # For backward compatibility with releases older than Quince.
+    from xblockutils.resources import ResourceLoader
+
+
+class AIEvalXBlock(StudioEditableXBlockMixin, XBlock):
+    """
+    Base class for Xblocks with AI evaluation
+    """
+
+    USER_KEY = "USER"
+    LLM_KEY = "LLM"
+
+    loader = ResourceLoader(__name__)
+
+    icon_class = "problem"
+    model_api_key = String(
+        display_name=_("Chosen model API Key"),
+        help=_("Enter your the API Key of your chosen model."),
+        default="",
+        scope=Scope.settings,
+    )
+    model_api_url = String(
+        display_name=_("Set your API URL"),
+        help=_(
+            "Fill this only for LLama. This required with models that don't have an official provider."
+            " Example URL: https://model-provider-example/llama3_70b"
+        ),
+        default=None,
+        scope=Scope.settings,
+    )
+    model = String(
+        display_name=_("AI model"),
+        help=_("Select the AI language model to use."),
+        values=[
+            {"display_name": model, "value": model} for model in SupportedModels.list()
+        ],
+        Scope=Scope.settings,
+        default=SupportedModels.GPT4O.value,
+    )
+
+    evaluation_prompt = String(
+        display_name=_("Evaluation prompt"),
+        help=_(
+            "Enter the evaluation prompt given to the model."
+            " The question will be inserted right after it."
+            " The student's answer would then follow the question. Markdown format can be used."
+        ),
+        default="You are a teacher. Evaluate the student's answer for the following question:",
+        multiline_editor=True,
+        scope=Scope.settings,
+    )
+    question = String(
+        display_name=_("Question"),
+        help=_(
+            "Enter the question you would like the students to answer."
+            " Markdown format can be used."
+        ),
+        default="",
+        multiline_editor=True,
+        scope=Scope.settings,
+    )
+
+    messages = Dict(
+        help=_("Dictionary with chat messages"),
+        scope=Scope.user_state,
+        default={USER_KEY: [], LLM_KEY: []},
+    )
+    editable_fields = (
+        "display_name",
+        "evaluation_prompt",
+        "question",
+        "model",
+        "model_api_key",
+        "model_api_url",
+    )
+
+    def resource_string(self, path):
+        """Handy helper for getting resources from our kit."""
+        data = pkg_resources.resource_string(__name__, path)
+        return data.decode("utf8")
+
+    def validate_field_data(self, validation, data):
+        """
+        Validate fields.
+        """
+
+        if not data.model or data.model not in SupportedModels.list():
+            validation.add(
+                ValidationMessage(
+                    ValidationMessage.ERROR,
+                    _(  # pylint: disable=translation-of-non-string
+                        f"Model field is mandatory and must be one of {', '.join(SupportedModels.list())}"
+                    ),
+                )
+            )
+
+        if not data.model_api_key:
+            validation.add(
+                ValidationMessage(
+                    ValidationMessage.ERROR, _("Model API key is mandatory")
+                )
+            )
+
+        if data.model == SupportedModels.LLAMA and not data.model_api_url:
+            validation.add(
+                ValidationMessage(
+                    ValidationMessage.ERROR,
+                    _("API URL field is mandatory when using ollama/llama2."),
+                )
+            )
+
+        if data.model != SupportedModels.LLAMA and data.model_api_url:
+            validation.add(
+                ValidationMessage(
+                    ValidationMessage.ERROR,
+                    _("API URL field can be set only when using ollama/llama2."),
+                )
+            )
+
+        if not data.question:
+            validation.add(
+                ValidationMessage(
+                    ValidationMessage.ERROR, _("Question field is mandatory")
+                )
+            )
diff --git a/ai_eval/coding_ai_eval.py b/ai_eval/coding_ai_eval.py
new file mode 100644
index 0000000..92f2658
--- /dev/null
+++ b/ai_eval/coding_ai_eval.py
@@ -0,0 +1,259 @@
+"""Coding Xblock with AI evaluation."""
+
+import logging
+import traceback
+import pkg_resources
+
+
+from django.utils.translation import gettext_noop as _
+from web_fragments.fragment import Fragment
+from xblock.core import XBlock
+from xblock.exceptions import JsonHandlerError
+from xblock.fields import Dict, Scope, String
+from xblock.validation import ValidationMessage
+
+from .llm import get_llm_response
+from .base import AIEvalXBlock
+from .utils import (
+    submit_code,
+    get_submission_result,
+    SUPPORTED_LANGUAGE_MAP,
+    LanguageLabels,
+)
+
+logger = logging.getLogger(__name__)
+
+USER_RESPONSE = "USER_RESPONSE"
+AI_EVALUATION = "AI_EVALUATION"
+CODE_EXEC_RESULT = "CODE_EXEC_RESULT"
+
+
+class CodingAIEvalXBlock(AIEvalXBlock):
+    """
+    TO-DO: document what your XBlock does.
+    """
+
+    has_author_view = True
+
+    display_name = String(
+        display_name=_("Display Name"),
+        help=_("Name of the component in the studio"),
+        default="Coding with AI Evaluation",
+        scope=Scope.settings,
+    )
+
+    judge0_api_key = String(
+        display_name=_("Judge0 API Key"),
+        help=_(
+            "Enter your the Judge0 API key used to execute code on Judge0."
+            " Get your key at https://rapidapi.com/judge0-official/api/judge0-ce."
+        ),
+        default="",
+        scope=Scope.settings,
+    )
+
+    language = String(
+        display_name=_("Programming Language"),
+        help=_("The programming language used for this Xblock."),
+        values=[
+            {"display_name": language, "value": language}
+            for language in SUPPORTED_LANGUAGE_MAP
+        ],
+        default=LanguageLabels.Python,
+        Scope=Scope.settings,
+    )
+    messages = Dict(
+        help=_("Dictionary with messages"),
+        scope=Scope.user_state,
+        default={USER_RESPONSE: "", AI_EVALUATION: "", CODE_EXEC_RESULT: {}},
+    )
+
+    editable_fields = AIEvalXBlock.editable_fields + ("judge0_api_key", "language")
+
+    def resource_string(self, path):
+        """Handy helper for getting resources from our kit."""
+        data = pkg_resources.resource_string(__name__, path)
+        return data.decode("utf8")
+
+    def student_view(self, context=None):
+        """
+        The primary view of the CodingAIEvalXBlock, shown to students
+        when viewing courses.
+        """
+        html = self.loader.render_django_template(
+            "/templates/coding_ai_eval.html",
+            {
+                "self": self,
+            },
+        )
+
+        frag = Fragment(html)
+        frag.add_css(self.resource_string("static/css/coding_ai_eval.css"))
+        frag.add_javascript(self.resource_string("static/js/src/utils.js"))
+
+        frag.add_javascript(self.resource_string("static/js/src/coding_ai_eval.js"))
+
+        monaco_html = self.loader.render_django_template(
+            "/templates/monaco.html",
+            {
+                "monaco_language": SUPPORTED_LANGUAGE_MAP[self.language].monaco_id,
+            },
+        )
+        marked_html = self.resource_string("static/html/marked-iframe.html")
+        js_data = {
+            "monaco_html": monaco_html,
+            "question": self.question,
+            "code": self.messages[USER_RESPONSE],
+            "ai_evaluation": self.messages[AI_EVALUATION],
+            "code_exec_result": self.messages[CODE_EXEC_RESULT],
+            "marked_html": marked_html,
+            "language": self.language,
+        }
+        frag.initialize_js("CodingAIEvalXBlock", js_data)
+        return frag
+
+    def author_view(self, context=None):
+        """
+        Create preview to be show to course authors in Studio.
+        """
+        if not self.validate():
+            fragment = Fragment()
+            fragment.add_content(
+                _(
+                    "To ensure this component works correctly, please fix the validation issues."
+                )
+            )
+            return fragment
+
+        return self.student_view(context=context)
+
+    def validate_field_data(self, validation, data):
+        """
+        Validate fields
+        """
+
+        super().validate_field_data(validation, data)
+
+        if data.language != LanguageLabels.HTML_CSS and not data.judge0_api_key:
+            validation.add(
+                ValidationMessage(
+                    ValidationMessage.ERROR, _("Judge0 API key is mandatory")
+                )
+            )
+
+    @XBlock.json_handler
+    def get_response(self, data, suffix=""):  # pylint: disable=unused-argument
+        """Get LLM feedback."""
+
+        answer = f"""
+        student code :
+
+        {data['code']}
+        """
+
+        # stdout and stderr only for executable languages (non HTML)
+        if self.language != LanguageLabels.HTML_CSS:
+            answer += f"""
+            stdout:
+
+            {data['stdout']}
+
+            stderr:
+
+            {data['stderr']}
+            """
+
+        messages = [
+            {
+                "role": "system",
+                "content": f"""
+               {self.evaluation_prompt}
+
+               {self.question}.
+
+               The programmimg language is {self.language}
+
+               Evaluation must be in Makrdown format.
+               """,
+            },
+            {
+                "content": f""" Here is the student's answer:
+              {answer}
+                """,
+                "role": "user",
+            },
+        ]
+
+        try:
+            response = get_llm_response(
+                self.model,
+                self.model_api_key,
+                messages,
+                self.model_api_url,
+            )
+
+        except Exception as e:
+            traceback.print_exc()
+            logger.error(
+                f"Failed while making LLM request using model {self.model}. Eaised error type: {type(e)}, Error: {e}"
+            )
+            raise JsonHandlerError(500, "A probem occured. Please retry.") from e
+
+        if response:
+            self.messages[USER_RESPONSE] = data["code"]
+            self.messages[AI_EVALUATION] = response
+            self.messages[CODE_EXEC_RESULT] = {
+                "stdout": data["stdout"],
+                "stderr": data["stderr"],
+            }
+            return {"response": response}
+
+        raise JsonHandlerError(500, "No AI Evaluation available. Please retry.")
+
+    @XBlock.json_handler
+    def submit_code_handler(self, data, suffix=""):  # pylint: disable=unused-argument
+        """
+        Submit code to Judge0.
+        """
+        submission_id = submit_code(
+            self.judge0_api_key, data["user_code"], self.language
+        )
+        return {"submission_id": submission_id}
+
+    @XBlock.json_handler
+    def reset_handler(self, data, suffix=""):  # pylint: disable=unused-argument
+        """
+        Reset the Xblock.
+        """
+        self.messages = {USER_RESPONSE: "", AI_EVALUATION: "", CODE_EXEC_RESULT: {}}
+        return {"message": "reset successful."}
+
+    @XBlock.json_handler
+    def get_submission_result_handler(
+        self, data, suffix=""
+    ):  # pylint: disable=unused-argument
+        """
+        Get code submission result.
+        """
+        submission_id = data["submission_id"]
+        return get_submission_result(self.judge0_api_key, submission_id)
+
+    @staticmethod
+    def workbench_scenarios():
+        """A canned scenario for display in the workbench."""
+        return [
+            (
+                "CodingAIEvalXBlock",
+                """<coding_ai_eval/>
+             """,
+            ),
+            (
+                "Multiple CodingAIEvalXBlock",
+                """<vertical_demo>
+                <coding_ai_eval/>
+                <coding_ai_eval/>
+                <coding_ai_eval/>
+                </vertical_demo>
+             """,
+            ),
+        ]
diff --git a/ai_eval/llm.py b/ai_eval/llm.py
new file mode 100644
index 0000000..0c956e7
--- /dev/null
+++ b/ai_eval/llm.py
@@ -0,0 +1,65 @@
+"""
+Integration with LLMs.
+"""
+
+from enum import Enum
+from litellm import completion
+
+
+class SupportedModels(Enum):
+    """
+    LLM Models supported by the CodingAIEvalXBlock and ShortAnswerAIEvalXBlock
+    """
+
+    GPT4O = "gpt-4o"
+    GEMINI_PRO = "gemini/gemini-pro"
+    CLAUDE_SONNET = "claude-3-5-sonnet-20240620"
+    LLAMA = "ollama/llama2"
+
+    @staticmethod
+    def list():
+        return [str(m.value) for m in SupportedModels]
+
+
+def get_llm_response(
+    model: SupportedModels, api_key: str, messages: list, api_base: str
+) -> str:
+    """
+    Get LLm response.
+
+    Args:
+        model (SupportedModels): The model to use for generating the response. This should be an instance of
+            the SupportedModels enum, specifying which LLM model to call.
+        api_key (str): The API key required for authenticating with the LLM service. This key should be kept
+            confidential and used to authorize requests to the service.
+        messages (list): A list of message objects to be sent to the LLM. Each message should be a dictionary
+            with the following format:
+
+            {
+                "content": str,   # The content of the message. This is the text that you want to send to the LLM.
+                "role": str       # The role of the message sender. This must be one of the following values:
+                                  # "user"    - Represents a user message.
+                                  # "system"  - Represents a system message, typically used for instructions or context.
+                                  # "assistant" - Represents a response or message from the LLM itself.
+            }
+
+            Example:
+            [
+                {"content": "Hello, how are you?", "role": "user"},
+                {"content": "I'm here to help you.", "role": "assistant"}
+            ]
+        api_base (str): The base URL of the LLM API endpoint. This is the root URL used to construct the full
+            API request URL. This is required only when using Llama which doesn't have an official provider.
+
+    Returns:
+        str: The response text from the LLM. This is typically the generated output based on the provided
+            messages.
+    """
+    kwargs = {}
+    if api_base:
+        kwargs["api_base"] = api_base
+    return (
+        completion(model=model, api_key=api_key, messages=messages, **kwargs)
+        .choices[0]
+        .message.content
+    )
diff --git a/ai_eval/shortanswer.py b/ai_eval/shortanswer.py
new file mode 100644
index 0000000..e9514da
--- /dev/null
+++ b/ai_eval/shortanswer.py
@@ -0,0 +1,152 @@
+"""Short answers Xblock with AI evaluation."""
+
+import logging
+import traceback
+
+
+from django.utils.translation import gettext_noop as _
+from web_fragments.fragment import Fragment
+from xblock.core import XBlock
+from xblock.exceptions import JsonHandlerError
+from xblock.fields import Integer, String, Scope
+from xblock.validation import ValidationMessage
+
+from .llm import get_llm_response
+from .base import AIEvalXBlock
+
+
+logger = logging.getLogger(__name__)
+
+
+class ShortAnswerAIEvalXBlock(AIEvalXBlock):
+    """
+    Short Answer Xblock.
+    """
+
+    display_name = String(
+        display_name=_("Display Name"),
+        help=_("Name of the component in the studio"),
+        default="Short answer with AI Evaluation",
+        scope=Scope.settings,
+    )
+
+    max_responses = Integer(
+        display_name=_("Max Responses"),
+        help=_("The maximum number of response messages the student can submit"),
+        scope=Scope.settings,
+        default=3,
+    )
+
+    editable_fields = AIEvalXBlock.editable_fields + ("max_responses",)
+
+    def validate_field_data(self, validation, data):
+        """
+        Validate fields
+        """
+
+        super().validate_field_data(validation, data)
+
+        if not data.max_responses or data.max_responses <= 0 or data.max_responses > 9:
+            validation.add(
+                ValidationMessage(
+                    ValidationMessage.ERROR,
+                    _("max responses must be an integer between 1 and 9"),
+                )
+            )
+
+    def student_view(self, context=None):
+        """
+        The primary view of the ShortAnswerAIEvalXBlock, shown to students
+        when viewing courses.
+        """
+
+        frag = Fragment()
+        frag.add_content(
+            self.loader.render_django_template(
+                "/templates/shortanswer.html",
+                {
+                    "self": self,
+                },
+            )
+        )
+
+        frag.add_css(self.resource_string("static/css/shortanswer.css"))
+        frag.add_javascript(self.resource_string("static/js/src/utils.js"))
+        frag.add_javascript(self.resource_string("static/js/src/shortanswer.js"))
+
+        marked_html = self.resource_string("static/html/marked-iframe.html")
+
+        js_data = {
+            "question": self.question,
+            "messages": self.messages,
+            "max_responses": self.max_responses,
+            "marked_html": marked_html,
+        }
+        frag.initialize_js("ShortAnswerAIEvalXBlock", js_data)
+        return frag
+
+    @XBlock.json_handler
+    def get_response(self, data, suffix=""):  # pylint: disable=unused-argument
+        """Get LLM feedback"""
+        user_submission = str(data["user_input"])
+        system_msg = {
+            "role": "system",
+            "content": f"""
+               {self.evaluation_prompt}
+
+               {self.question}.
+
+               Evaluation must be in Makrdown format.
+               """,
+        }
+        messages = [system_msg]
+        # add previous messages
+        # the first AI role is 'system' which defines the LLM's personnality and behavior.
+        # subsequent roles are 'assistant' and 'user'
+        for i in range(len(self.messages[self.USER_KEY])):
+            messages.append(
+                {"content": self.messages[self.USER_KEY][i], "role": "user"}
+            )
+            messages.append(
+                {"content": self.messages[self.LLM_KEY][i], "role": "assistant"}
+            )
+
+        messages.append({"role": "user", "content": user_submission})
+
+        try:
+            response = get_llm_response(
+                self.model, self.model_api_key, messages, self.model_api_url
+            )
+
+        except Exception as e:
+            traceback.print_exc()
+            logger.error(
+                f"Failed while making LLM request using model {self.model}. Eaised error type: {type(e)}, Error: {e}"
+            )
+            raise JsonHandlerError(500, "A probem occured. Please retry.") from e
+
+        if response:
+            self.messages[self.USER_KEY].append(user_submission)
+            self.messages[self.LLM_KEY].append(response)
+            return {"response": response}
+
+        raise JsonHandlerError(500, "A probem occured. The LLM sent an empty response.")
+
+    @staticmethod
+    def workbench_scenarios():
+        """A canned scenario for display in the workbench."""
+        return [
+            (
+                "ShortAnswerAIEvalXBlock",
+                """<shortanswer/>
+             """,
+            ),
+            (
+                "Multiple ShortAnswerAIEvalXBlock",
+                """<vertical_demo>
+                <shortanswer/>
+                <shortanswer/>
+                </vertical_demo>
+             """,
+            ),
+        ]
diff --git a/ai_eval/static/README.txt b/ai_eval/static/README.txt
new file mode 100644
index 0000000..0472ef6
--- /dev/null
+++ b/ai_eval/static/README.txt
@@ -0,0 +1,19 @@
+This static directory is for files that should be included in your kit as plain
+static files.
+
+You can ask the runtime for a URL that will retrieve these files with:
+
+    url = self.runtime.local_resource_url(self, "static/js/lib.js")
+
+The default implementation is very strict though, and will not serve files from
+the static directory.  It will serve files from a directory named "public".
+Create a directory alongside this one named "public", and put files there.
+Then you can get a url with code like this:
+
+    url = self.runtime.local_resource_url(self, "public/js/lib.js")
+
+The sample code includes a function you can use to read the content of files
+in the static directory, like this:
+
+    frag.add_javascript(self.resource_string("static/js/my_block.js"))
+
diff --git a/ai_eval/static/css/coding_ai_eval.css b/ai_eval/static/css/coding_ai_eval.css
new file mode 100644
index 0000000..1dc3548
--- /dev/null
+++ b/ai_eval/static/css/coding_ai_eval.css
@@ -0,0 +1,126 @@
+.used-prog-language {
+  flex: 5;
+  margin: 0.5rem 1rem;
+  text-align: end;
+}
+
+.used-prog-language span {
+  background-color: #00262b;
+  color: white;
+  display: inline-block;
+  padding: .25em .4em;
+  font-size: 75%;
+  font-weight: 700;
+  line-height: 1;
+  text-align: center;
+  white-space: nowrap;
+  vertical-align: baseline;
+  border-radius: .25rem;
+}
+
+.eval-ai-container {
+  display: flex;
+  max-height: 600px;
+  box-shadow: 0px 0px 10px 0px #0000001A;
+  border: 1px solid #CED4DA;
+  border-radius: 4px;
+}
+
+.eval-ai-container .eval-ai-code-editor {
+  flex: 5;
+}
+
+.eval-ai-container #monaco {
+  min-height: 500px;
+  width: 100%;
+  border-bottom: 1.5px solid #CED4DA;
+}
+
+.eval-ai-container .result {
+  flex: 3;
+  background-color: #f5f5f5;
+  overflow-y: hidden;
+  border-top-right-radius: 4px;
+  border-bottom-right-radius: 4px;
+}
+
+.eval-ai-container .html-render {
+  height: 100vh;
+}
+
+.eval-ai-container .eval-ai-buttons {
+  display: flex;
+  justify-content: flex-end;
+}
+
+.eval-ai-container .submit-loader {
+  margin-left: 5px;
+}
+
+.eval-ai-container .eval-ai-buttons .eval-ai-button {
+  margin: 10px;
+  text-align: center;
+  height: fit-content;
+  padding: 8px 12px;
+  border-radius: 4px;
+  font-size: 0.875em;
+  line-height: 1.25em;
+}
+
+.eval-ai-container .eval-ai-buttons .disabled-btn {
+  opacity: 0.8;
+  cursor: not-allowed !important;
+}
+
+.eval-ai-container .stderr {
+  margin-top: 10px;
+  color: red;
+  white-space: pre-wrap;
+}
+
+.eval-ai-container .stdout {
+  white-space: pre-wrap;
+}
+
+/* tabs */
+.tab {
+  display: flex;
+  border-bottom: 1px solid #edeae9;
+}
+
+.tab .result-tab-btn {
+  float: left;
+  border: none;
+  outline: none;
+  cursor: pointer;
+  padding: 8px 12px;
+  transition: 0.3s;
+  font-size: 0.875em;
+  line-height: 1.25em;
+}
+
+.tab .result-tab-btn:hover {
+  background-color: #ddd;
+}
+
+.tab .result-tab-btn.active {
+  border-bottom: 2px solid #546d70;
+}
+
+.tabcontent {
+  padding: 6px 12px;
+  border-top: none;
+  animation: fadeEffect 1s;
+  overflow-y: scroll;
+  height: calc(100% - 48px);
+}
+
+@keyframes fadeEffect {
+  from {
+    opacity: 0;
+  }
+
+  to {
+    opacity: 1;
+  }
+}
\ No newline at end of file
diff --git a/ai_eval/static/css/shortanswer.css b/ai_eval/static/css/shortanswer.css
new file mode 100644
index 0000000..48ab5b8
--- /dev/null
+++ b/ai_eval/static/css/shortanswer.css
@@ -0,0 +1,127 @@
+/* CSS for ShortAnswerAIEvalXBlock */
+
+.shortanswer_block .count {
+  font-weight: bold;
+}
+
+.shortanswer_block p {
+  cursor: pointer;
+}
+
+.chat-history {
+  min-height: 250px;
+  max-height: 400px;
+  overflow-y: auto;
+  margin-bottom: 10px;
+  padding: 0px 2em;
+}
+
+.chat-message-container {
+  display: flex;
+}
+.chat-message-container .chat-message {
+  flex: 0 1 auto;
+  max-width: 80%;
+  padding: 10px 10px 5px 10px;
+  margin-bottom: 10px;
+  border-radius: 5px;
+}
+
+.user-answer {
+  text-align: right;
+  margin-left: auto;
+  background-color: #f0f3f5;
+  color: #44586c;
+}
+
+.ai-eval,
+.message-spinner {
+  text-align: left;
+  color: white;
+  margin-right: auto;
+  background-color: #476480;
+}
+.ai-eval * {
+  color: white !important;
+}
+
+.submit-row {
+  display: flex;
+  margin-bottom: 1em;
+  font-size: 16px;
+  padding: 10px;
+}
+
+.submit-row .user-input {
+  flex: 8;
+  height: auto;
+  margin-right: 10px;
+  border: 1px solid gray;
+  border-radius: 5px;
+  overflow: auto;
+  max-height: 200px;
+  resize: none;
+}
+
+.submit-row #submit-button {
+  flex: 1;
+  height: fit-content;
+  text-align: center;
+  padding: 10px 0;
+  margin-left: auto;
+  background-color: #00262b;
+  color: white;
+  border-radius: 5px;
+  border: none;
+  cursor: pointer;
+}
+.submit-row .disabled-btn {
+  opacity: 0.8;
+  cursor: not-allowed !important;
+}
+/* spinner animation */
+.chat-message-container .message-spinner > div {
+  width: 4px;
+  height: 4px;
+  margin-right: 2px;
+  background-color: white;
+
+  border-radius: 100%;
+  display: inline-block;
+  -webkit-animation: chat-block-sk-bouncedelay 1.4s infinite ease-in-out both;
+  animation: chat-block-sk-bouncedelay 1.4s infinite ease-in-out both;
+}
+
+.chat-message-container.message-spinner .bounce1 {
+  -webkit-animation-delay: -0.32s;
+  animation-delay: -0.32s;
+}
+
+.chat-message-container .message-spinner .bounce2 {
+  -webkit-animation-delay: -0.16s;
+  animation-delay: -0.16s;
+}
+
+@-webkit-keyframes chat-block-sk-bouncedelay {
+  0%,
+  80%,
+  100% {
+    -webkit-transform: scale(0);
+  }
+  40% {
+    -webkit-transform: scale(1);
+  }
+}
+
+@keyframes chat-block-sk-bouncedelay {
+  0%,
+  80%,
+  100% {
+    -webkit-transform: scale(0);
+    transform: scale(0);
+  }
+  40% {
+    -webkit-transform: scale(1);
+    transform: scale(1);
+  }
+}
diff --git a/ai_eval/static/html/marked-iframe.html b/ai_eval/static/html/marked-iframe.html
new file mode 100644
index 0000000..d21a301
--- /dev/null
+++ b/ai_eval/static/html/marked-iframe.html
@@ -0,0 +1,7 @@
+<!doctype html>
+<html lang="en">
+<head></head>
+<body>
+    <script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/marked/13.0.2/marked.min.js"></script>
+</body>
+</html>
\ No newline at end of file
diff --git a/ai_eval/static/js/src/coding_ai_eval.js b/ai_eval/static/js/src/coding_ai_eval.js
new file mode 100644
index 0000000..e5ef4d5
--- /dev/null
+++ b/ai_eval/static/js/src/coding_ai_eval.js
@@ -0,0 +1,235 @@
+/* Javascript for CodingAIEvalXBlock. */
+function CodingAIEvalXBlock(runtime, element, data) {
+  const runCodeHandlerURL = runtime.handlerUrl(element, "submit_code_handler");
+  const resetHandlerURL = runtime.handlerUrl(element, "reset_handler");
+
+
+  const submissionResultURL = runtime.handlerUrl(
+    element,
+    "get_submission_result_handler",
+  );
+  loadMarkedInIframe(data.marked_html);
+  const llmResponseHandlerURL = runtime.handlerUrl(element, "get_response");
+  const HTML_CSS = "HTML/CSS";
+  const HTML_PLACEHOLER =
+    "<!DOCTYPE html>\n<html>\n<head>\n<style>\nbody {background: linear-gradient(90deg, #ffecd2, #fcb69f);}\nh1   {font-style: italic;}\np    {border: 2px solid powderblue;}\n</style>\n</head>\n<body>\n<h1>This is a heading</h1>\n<p>This is a paragraph.</p>\n</body>\n</html>";
+
+  const iframe = $("#monaco", element)[0];
+  const submitButton = $("#submit-button", element);
+  const resetButton = $("#reset-button", element);
+  const AIFeeback = $("#ai-feedback", element);
+  const stdout = $(".stdout", element);
+  const stderr = $(".stderr", element);
+  const htmlRenderIframe = $(".html-render", element);
+
+  const MAX_JUDGE0_RETRY_ITER = 5;
+  const WAIT_TIME_MS = 1000;
+
+  $(function () {
+    // The newer runtime uses the 'data-usage' attribute, while the LMS uses 'data-usage-id'
+    // A Jquery object can sometimes be returned e.g. after a studio field edit, we handle it with ?.[0]
+    const xblockUsageId =
+      element.getAttribute?.("data-usage") ||
+      element.getAttribute?.("data-usage-id") ||
+      element?.[0].getAttribute("data-usage-id");
+
+    if (!xblockUsageId) {
+      throw new Error(
+        "XBlock is missing a usage ID attribute on its root HTML node.",
+      );
+    }
+
+    // __USAGE_ID_PLACEHOLDER__ is the event data sent from the monaco iframe after loading
+    // we rely on the usage_id to limit the event to the Xblock scope
+    iframe.srcdoc = data.monaco_html.replace(
+      "__USAGE_ID_PLACEHOLDER__",
+      xblockUsageId,
+    );
+    runFuncAfterLoading(init);
+    function submitCode() {
+      const code = iframe.contentWindow.editor.getValue();
+      return $.ajax({
+        url: runCodeHandlerURL,
+        method: "POST",
+        data: JSON.stringify({ user_code: code }),
+      });
+    }
+    function delay(ms, data) {
+      const deferred = $.Deferred();
+      setTimeout(function () {
+        deferred.resolve(data);
+      }, ms);
+      return deferred.promise();
+    }
+    function getSubmissionResult(data) {
+      let retries = 0;
+      const deferred = $.Deferred();
+
+      function attempt() {
+        return $.ajax({
+          url: submissionResultURL,
+          method: "POST",
+          data: JSON.stringify({ submission_id: data.submission_id }),
+        })
+          .then(function (result) {
+            console.log("result", result, retries)
+            if (result.status.id === 1 || result.status.id === 2) {
+              // https://ce.judge0.com/#statuses-and-languages-status-get 
+              // Retry if status is 1 (In Queue) or 2 (Processing)
+              if (retries < MAX_JUDGE0_RETRY_ITER) {
+                retries++;
+                console.log("Judge0 fetch result retry attempt:", retries);
+                setTimeout(function () {
+                  attempt();
+                }, WAIT_TIME_MS);
+              } else {
+                deferred.reject(new Error("Judge0 submission result fetch failed after " + MAX_JUDGE0_RETRY_ITER + " attempts."));
+              }
+            } else {
+              const output = [result.compile_output, result.stdout].join("\n").trim();
+              stdout.text(output);
+              stderr.text(result.stderr);
+              deferred.resolve(result);
+            }
+
+
+          })
+          .fail(function (error) {
+            console.log("Error: ", error);
+            deferred.reject(new Error("An error occured while trying to fetch Judge0 submission result."));
+          });
+      }
+      attempt();
+      return deferred.promise();
+    }
+
+    function getLLMFeedback(data) {
+      return $.ajax({
+        url: llmResponseHandlerURL,
+        method: "POST",
+        data: JSON.stringify({
+          code: iframe.contentWindow.editor.getValue(),
+          stdout: data.stdout,
+          stderr: data.stderr,
+        }),
+        success: function (data) {
+          console.log(data);
+          AIFeeback.html(MarkdownToHTML(data.response));
+          $("#ai-feedback-tab", element).click();
+        },
+      });
+    }
+
+    resetButton.click(() => {
+      $.ajax({
+        url: resetHandlerURL,
+        method: "POST",
+        data: JSON.stringify({}),
+        success: function (data) {
+          iframe.contentWindow.editor.setValue("");
+          AIFeeback.html("");
+          if (data.language !== HTML_CSS) {
+            stdout.text("");
+            stderr.text("");
+          }
+        },
+        error: function(xhr, status, error) {
+          console.error('Error:', error);
+          alert("A problem occured during reset.");
+        }
+      });
+
+    });
+
+    submitButton.click(() => {
+      const code = iframe.contentWindow.editor.getValue();
+      if (!code?.length) {
+        return;
+      }
+      disableSubmitButton();
+      var deferred = null;
+      if (data.language === HTML_CSS) {
+        // no need to submit HTML code, we directly get LLM feedback
+        deferred = getLLMFeedback({ stdout: "", stderr: "" });
+      } else {
+        deferred = submitCode()
+          .then(function (data) {
+            return delay(WAIT_TIME_MS * 2, data);
+          })
+          .then(getSubmissionResult)
+          .then(getLLMFeedback);
+      }
+
+      deferred
+        .done(function (data) {
+          enableSubmitButton();
+        })
+        .fail(function (error) {
+          console.log("Error: ", error);
+          enableSubmitButton();
+          alert("A problem occured while submitting the code.");
+        });
+    });
+
+    function init() {
+      $("#question-text", element).html(MarkdownToHTML(data.question));
+      // Triggered when the Monaco editor loads.
+      // Since a Unit can have multiple instances of this Xblock,
+      // we use the XblockUsageId to differentiate between them.
+      window.addEventListener("message", function (event) {
+        if (event.data === xblockUsageId) {
+          if (data.code?.length) {
+            iframe.contentWindow.editor.setValue(data.code);
+          }
+
+          AIFeeback.html(MarkdownToHTML(data.ai_evaluation || ""));
+          if (data.language === HTML_CSS) {
+            // render HTML/CSS into iframe
+            if (data.code?.length) {
+              renderUserHTML(data.code);
+            } else {
+              iframe.contentWindow.editor.setValue(HTML_PLACEHOLER);
+              renderUserHTML(HTML_PLACEHOLER);
+            }
+            addMonacoHTMLRenderEventListener();
+          } else {
+            // load existing results for executable languages
+            stdout.text(data.code_exec_result?.stdout || "");
+            stderr.text(data.code_exec_result?.stderr || "");
+          }
+        }
+      });
+    }
+    function addMonacoHTMLRenderEventListener() {
+      iframe.contentWindow.editor.onDidChangeModelContent((event) => {
+        renderUserHTML(iframe.contentWindow.editor.getValue());
+      });
+    }
+    function renderUserHTML(userHTML) {
+      htmlRenderIframe.attr("srcdoc", stripScriptTags(userHTML));
+    }
+  });
+
+  function disableSubmitButton() {
+    submitButton.append('<i class="fa fa-spinner fa-spin submit-loader"></i>');
+    submitButton.prop("disabled", true);
+    submitButton.addClass("disabled-btn");
+  }
+
+  function enableSubmitButton() {
+    $(".submit-loader", element).remove();
+    submitButton.prop("disabled", false);
+    submitButton.removeClass("disabled-btn");
+  }
+
+  // basic tabs
+  $(".tablinks", element).click((event) => {
+    $(".tabcontent", element).hide();
+    $(".tablinks", element).removeClass("active");
+    $(event.target).addClass("active");
+    const contentID = "#" + $(event.target).data("id");
+    $(contentID, element).show();
+  });
+  // default tab
+  $("#defaultOpen", element).click();
+}
diff --git a/ai_eval/static/js/src/shortanswer.js b/ai_eval/static/js/src/shortanswer.js
new file mode 100644
index 0000000..4bbd0ab
--- /dev/null
+++ b/ai_eval/static/js/src/shortanswer.js
@@ -0,0 +1,105 @@
+/* Javascript for ShortAnswerAIEvalXBlock. */
+function ShortAnswerAIEvalXBlock(runtime, element, data) {
+  const handlerUrl = runtime.handlerUrl(element, "get_response");
+
+  loadMarkedInIframe(data.marked_html);
+
+  $(function () {
+    const spinner = $(".message-spinner", element);
+    const spinnnerContainer = $("#chat-spinner-container", element);
+    const submitButton = $("#submit-button", element);
+    const userInput = $(".user-input", element);
+    const userInputElem = userInput[0];
+    let initDone = false;
+
+    runFuncAfterLoading(init);
+
+    function getResponse() {
+      if (!userInput.val().length) return;
+
+      disableInput();
+      spinner.show();
+      insertUserMessage(userInput.val());
+      $.ajax({
+        url: handlerUrl,
+        method: "POST",
+        data: JSON.stringify({ user_input: userInput.val() }),
+        success: function (response) {
+          spinner.hide();
+          insertAIMessage(response.response);
+          userInput.val("");
+          if ($(".user-answer", element).length >= data.max_responses) {
+            disableInput();
+          } else {
+            enableInput();
+          }
+        },
+        error: function (jqXHR, textStatus, errorThrown) {
+          spinner.hide();
+          alert(errorThrown);
+
+          deleteLastMessage();
+          enableInput();
+        },
+      });
+    }
+
+    submitButton.click(getResponse);
+
+    function disableInput() {
+      userInput.prop("disabled", true);
+      userInput.removeAttr("placeholder");
+      submitButton.prop("disabled", true);
+      submitButton.addClass("disabled-btn");
+    }
+
+    function enableInput() {
+      userInput.prop("disabled", false);
+      submitButton.prop("disabled", false);
+      submitButton.removeClass("disabled-btn");
+    }
+
+    function adjustTextareaHeight(element) {
+      element.style.height = "";
+      element.style.height = element.scrollHeight + "px";
+    }
+    userInputElem.addEventListener("input", (event) => {
+      adjustTextareaHeight(userInputElem);
+    });
+
+    function init() {
+      if (initDone) return;
+      initDone = true;
+      $("#question-text", element).html(MarkdownToHTML(data.question));
+      for (let i = 0; i < data.messages.USER.length; i++) {
+        insertUserMessage(data.messages.USER[i]);
+        insertAIMessage(data.messages.LLM[i]);
+      }
+      if (
+        data.messages.USER.length &&
+        data.messages.USER.length >= data.max_responses
+      ) {
+        disableInput();
+      }
+    }
+
+    function insertUserMessage(msg) {
+      if (msg?.length) {
+        $(` <div class="chat-message-container">
+                <div class="chat-message user-answer">${MarkdownToHTML(msg)}</div>
+      </div>`).insertBefore(spinnnerContainer);
+      }
+    }
+
+    function insertAIMessage(msg) {
+      if (msg?.length) {
+        $(` <div class="chat-message-container">
+                <div class="chat-message ai-eval">${MarkdownToHTML(msg)}</div>
+      </div>`).insertBefore(spinnnerContainer);
+      }
+    }
+    function deleteLastMessage() {
+      spinnnerContainer.prev().remove();
+    }
+  });
+}
diff --git a/ai_eval/static/js/src/utils.js b/ai_eval/static/js/src/utils.js
new file mode 100644
index 0000000..063ef98
--- /dev/null
+++ b/ai_eval/static/js/src/utils.js
@@ -0,0 +1,41 @@
+// we load Marked in a Iframe to avoid RequireJS conflicts with the Xblock runtime
+function loadMarkedInIframe(html) {
+  if (!window.marked && !document.getElementById("marked-iframe")) {
+    const iframe = document.createElement("iframe");
+    iframe.style.display = "none";
+    iframe.setAttribute("id", "marked-iframe");
+    iframe.srcdoc = html;
+    document.body.appendChild(iframe);
+  }
+}
+function MarkdownToHTML(text) {
+  window.marked =
+    window.marked ??
+    document.getElementById("marked-iframe").contentWindow.marked;
+  if (typeof marked !== "undefined") text = marked.parse(text);
+  return text;
+}
+
+function runFuncAfterLoading(func) {
+  const markedIframe = document.getElementById("marked-iframe");
+  // loading is finished when Marked is available
+  if (markedIframe.contentWindow.marked) {
+    func();
+  } else {
+    markedIframe.addEventListener("load", (event) => {
+      func();
+    });
+  }
+}
+
+function stripScriptTags(html) {
+  const div = document.createElement("div");
+  div.innerHTML = html;
+
+  const scripts = div.getElementsByTagName("script");
+  while (scripts.length > 0) {
+    scripts[0].parentNode.removeChild(scripts[0]);
+  }
+
+  return div.innerHTML;
+}
diff --git a/ai_eval/templates/coding_ai_eval.html b/ai_eval/templates/coding_ai_eval.html
new file mode 100644
index 0000000..4472024
--- /dev/null
+++ b/ai_eval/templates/coding_ai_eval.html
@@ -0,0 +1,31 @@
+<div id="question-text"></div>
+<div class="eval-ai-container">
+  <div class="eval-ai-code-editor">
+    <div class="used-prog-language">
+      <span>{{ self.language }}</span>
+    </div>
+    <iframe id="monaco" frameborder="0" scrolling="no"></iframe>
+    <div class="eval-ai-buttons">
+      <button id="reset-button" class="eval-ai-button">Reset</button>
+      <button id="submit-button" class="eval-ai-button btn btn-primary">Submit Code</button>
+    </div>
+  </div>
+
+  <div class="result">
+    <div class="tab">
+      <span class="tablinks result-tab-btn" data-id="output" id="defaultOpen">Output</span>
+      <span class="tablinks result-tab-btn" data-id="ai-feedback" id="ai-feedback-tab">AI feedback</span>
+    </div>
+
+    <div id="output" class="tabcontent">
+      {% if self.language == "HTML/CSS" %}
+      <iframe class="html-render" scrolling="no" frameBorder="0"></iframe>
+      {% else %}
+      <div class="stdout"></div>
+      <div class="stderr"></div>
+      {% endif %}
+    </div>
+
+    <div id="ai-feedback" class="tabcontent"></div>
+  </div>
+</div>
diff --git a/ai_eval/templates/monaco.html b/ai_eval/templates/monaco.html
new file mode 100644
index 0000000..56de7e5
--- /dev/null
+++ b/ai_eval/templates/monaco.html
@@ -0,0 +1,22 @@
+<!doctype html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <title>Monaco Editor</title>
+</head>
+<body style="margin: 0;">
+  <div class="monaco" style="width: 100vw; height: 100vh;"></div>
+  <script type="module">
+    import * as monaco from "https://cdn.jsdelivr.net/npm/monaco-editor@0.49.0/+esm";
+
+    window.editor = monaco.editor.create(document.querySelector(".monaco"), {
+      language: "{{ monaco_language }}",
+      minimap: { enabled: false },
+      lineNumbersMinChars: 2,
+      folding: false,
+    });
+
+    window.parent.postMessage('__USAGE_ID_PLACEHOLDER__', '*');
+  </script>
+</body>
+</html>
diff --git a/ai_eval/templates/shortanswer.html b/ai_eval/templates/shortanswer.html
new file mode 100644
index 0000000..cc73600
--- /dev/null
+++ b/ai_eval/templates/shortanswer.html
@@ -0,0 +1,26 @@
+<div class="shortanswer_block">
+    <div>
+        <div id="question-text">
+            Loading ..
+        </div>
+    </div>
+
+    <div>
+        <div class="chat-history">
+            <div class="chat-message-container" id="chat-spinner-container">
+                <div class="chat-message message-spinner" style="display: none;">
+                    <div class="bounce1">
+                    </div>
+                    <div class="bounce2">
+                    </div>
+                    <div class="bounce3">
+                    </div>
+                </div>
+            </div>
+        </div>
+        <div class="submit-row">
+            <textarea class="user-input" rows="1" placeholder="Type your answer here" maxlength="1000"></textarea>
+            <span id="submit-button">Submit <i class="fa fa-paper-plane"></i></span>
+        </div>
+    </div>
+</div>
\ No newline at end of file
diff --git a/ai_eval/tests/__init__.py b/ai_eval/tests/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/ai_eval/tests/test_ai_eval.py b/ai_eval/tests/test_ai_eval.py
new file mode 100644
index 0000000..0623558
--- /dev/null
+++ b/ai_eval/tests/test_ai_eval.py
@@ -0,0 +1,55 @@
+"""
+Testing module.
+"""
+
+import unittest
+from xblock.field_data import DictFieldData
+from xblock.test.toy_runtime import ToyRuntime
+from ai_eval import CodingAIEvalXBlock, ShortAnswerAIEvalXBlock
+
+
+class TestCodingAIEvalXBlock(unittest.TestCase):
+    """Tests for CodingAIEvalXBlock"""
+
+    def test_basics_student_view(self):
+        """Test the basic view loads."""
+        data = {
+            "language": "Python",
+            "question": "ca va?",
+            "code": "",
+            "ai_evaluation": "",
+            "code_exec_result": {},
+            "marked_html": '<!doctype html>\n<html lang="en">\n<head></head>\n<body>\n    <script '
+            'type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/marked/13.0.2/marked'
+            '.min.js"></script>\n</body>\n</html>',
+            "monaco_html": '<!doctype html>\n<html lang="en">\n<head>\n  <meta charset="UTF-8" />\n'
+            '  <title>Monaco Editor</title>\n</head>\n<body style="margin: 0;">\n  <div class="monaco"'
+            ' style="width: 100vw; height: 100vh;"></div>\n  <script type="module">\n    import * as'
+            ' monaco from "https://cdn.jsdelivr.net/npm/monaco-editor@0.49.0/+esm";\n\n    window.editor '
+            '= monaco.editor.create(document.querySelector(".monaco"), {\n      language: "python",\n    '
+            '  minimap: { enabled: false },\n      lineNumbersMinChars: 2,\n      folding: false,\n    });\n\n'
+            '    window.parent.postMessage(\'__USAGE_ID_PLACEHOLDER__\', \'*\');\n  </script>\n</body>\n</html>\n',
+        }
+        block = CodingAIEvalXBlock(ToyRuntime(), DictFieldData(data), None)
+        frag = block.student_view()
+        self.assertEqual(data, frag.json_init_args)
+        self.assertIn('<div class="eval-ai-container">', frag.content)
+
+
+class TestShortAnswerAIEvalXBlock(unittest.TestCase):
+    """Tests for ShortAnswerAIEvalXBlock"""
+
+    def test_basics_student_view(self):
+        """Test the basic view loads."""
+        data = {
+            "question": "ca va?",
+            "messages": {"USER": [], "LLM": []},
+            "max_responses": 3,
+            "marked_html": '<!doctype html>\n<html lang="en">\n<head></head>\n<body>\n    <script '
+            'type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/marked/13.0.2/marked'
+            '.min.js"></script>\n</body>\n</html>',
+        }
+        block = ShortAnswerAIEvalXBlock(ToyRuntime(), DictFieldData(data), None)
+        frag = block.student_view()
+        self.assertEqual(data, frag.json_init_args)
+        self.assertIn('<div class="shortanswer_block">', frag.content)
diff --git a/ai_eval/translations/README.txt b/ai_eval/translations/README.txt
new file mode 100644
index 0000000..0493bcc
--- /dev/null
+++ b/ai_eval/translations/README.txt
@@ -0,0 +1,4 @@
+Use this translations directory to provide internationalized strings for your XBlock project.
+
+For more information on how to enable translations, visit the Open edX XBlock tutorial on Internationalization:
+http://edx.readthedocs.org/projects/xblock-tutorial/en/latest/edx_platform/edx_lms.html
diff --git a/ai_eval/utils.py b/ai_eval/utils.py
new file mode 100644
index 0000000..cf866bf
--- /dev/null
+++ b/ai_eval/utils.py
@@ -0,0 +1,84 @@
+"""
+Utilities
+"""
+
+from dataclasses import dataclass
+import requests
+
+
+@dataclass
+class ProgrammimgLanguage:
+    """A programming language."""
+
+    monaco_id: str
+    judge0_id: int
+
+
+class LanguageLabels:
+    """Language labels as seen by users."""
+
+    Python = "Python"
+    JavaScript = "JavaScript"
+    Java = "Java"
+    CPP = "C++"
+    HTML_CSS = "HTML/CSS"
+
+
+# supported programming languages and their IDs in judge0 and monaco
+# https://ce.judge0.com/#statuses-and-languages-active-and-archived-languages
+SUPPORTED_LANGUAGE_MAP = {
+    LanguageLabels.Python: ProgrammimgLanguage(
+        monaco_id="python", judge0_id=92
+    ),  # Python (3.11.2)
+    LanguageLabels.JavaScript: ProgrammimgLanguage(
+        monaco_id="javascript", judge0_id=93
+    ),  # JavaScript (Node.js 18.15.0)
+    LanguageLabels.Java: ProgrammimgLanguage(
+        monaco_id="java", judge0_id=91
+    ),  # Java (JDK 17.0.6)
+    LanguageLabels.CPP: ProgrammimgLanguage(
+        monaco_id="cpp", judge0_id=54
+    ),  # C++ (GCC 9.2.0)
+    # Monaco's HTML support includes CSS support within the 'style' tag.
+    LanguageLabels.HTML_CSS: ProgrammimgLanguage(
+        monaco_id="html", judge0_id=-1
+    ),  # no exec
+}
+
+
+JUDGE0_BASE_CE_URL = "https://judge0-ce.p.rapidapi.com"
+
+
+def submit_code(api_key: str, code: str, language: str) -> str:
+    """
+    Submit code to the judge0 API.
+    """
+    url = f"{JUDGE0_BASE_CE_URL}/submissions?base64_encoded=false&wait=false"
+    headers = {"content-type": "application/json", "x-rapidapi-key": api_key}
+
+    data = {
+        "source_code": code,
+        "language_id": SUPPORTED_LANGUAGE_MAP[language].judge0_id,
+    }
+
+    response = requests.post(url, headers=headers, json=data, timeout=10)
+    response.raise_for_status()
+    result = response.json()
+    sub_id = result["token"]
+
+    return sub_id
+
+
+def get_submission_result(api_key: str, submission_id: str):
+    """
+    Get result from Judge0 submission.
+    """
+
+    url = f"{JUDGE0_BASE_CE_URL}/submissions/{submission_id}?base64_encoded=false&fields=*"
+    headers = {"content-type": "application/json", "x-rapidapi-key": api_key}
+
+    response = requests.get(url, headers=headers, timeout=10)
+    response.raise_for_status()
+    result = response.json()
+
+    return result
diff --git a/docs/coding-xblock-ai-feedback.png b/docs/coding-xblock-ai-feedback.png
new file mode 100644
index 0000000..d676db2
Binary files /dev/null and b/docs/coding-xblock-ai-feedback.png differ
diff --git a/docs/coding-xblock-html.png b/docs/coding-xblock-html.png
new file mode 100644
index 0000000..df43202
Binary files /dev/null and b/docs/coding-xblock-html.png differ
diff --git a/docs/coding-xblock.png b/docs/coding-xblock.png
new file mode 100644
index 0000000..cf8a0a3
Binary files /dev/null and b/docs/coding-xblock.png differ
diff --git a/docs/shortanswer-xblock.png b/docs/shortanswer-xblock.png
new file mode 100644
index 0000000..b55f91f
Binary files /dev/null and b/docs/shortanswer-xblock.png differ
diff --git a/pylintrc b/pylintrc
new file mode 100644
index 0000000..2c17a7b
--- /dev/null
+++ b/pylintrc
@@ -0,0 +1,389 @@
+# ***************************
+# ** DO NOT EDIT THIS FILE **
+# ***************************
+#
+# This file was generated by edx-lint: https://github.com/openedx/edx-lint
+#
+# If you want to change this file, you have two choices, depending on whether
+# you want to make a local change that applies only to this repo, or whether
+# you want to make a central change that applies to all repos using edx-lint.
+#
+# Note: If your pylintrc file is simply out-of-date relative to the latest
+# pylintrc in edx-lint, ensure you have the latest edx-lint installed
+# and then follow the steps for a "LOCAL CHANGE".
+#
+# LOCAL CHANGE:
+#
+# 1. Edit the local pylintrc_tweaks file to add changes just to this
+#    repo's file.
+#
+# 2. Run:
+#
+#       $ edx_lint write pylintrc
+#
+# 3. This will modify the local file.  Submit a pull request to get it
+#    checked in so that others will benefit.
+#
+#
+# CENTRAL CHANGE:
+#
+# 1. Edit the pylintrc file in the edx-lint repo at
+#    https://github.com/openedx/edx-lint/blob/master/edx_lint/files/pylintrc
+#
+# 2. install the updated version of edx-lint (in edx-lint):
+#
+#       $ pip install .
+#
+# 3. Run (in edx-lint):
+#
+#       $ edx_lint write pylintrc
+#
+# 4. Make a new version of edx_lint, submit and review a pull request with the
+#    pylintrc update, and after merging, update the edx-lint version and
+#    publish the new version.
+#
+# 5. In your local repo, install the newer version of edx-lint.
+#
+# 6. Run:
+#
+#       $ edx_lint write pylintrc
+#
+# 7. This will modify the local file.  Submit a pull request to get it
+#    checked in so that others will benefit.
+#
+#
+#
+#
+#
+# STAY AWAY FROM THIS FILE!
+#
+#
+#
+#
+#
+# SERIOUSLY.
+#
+# ------------------------------
+# Generated by edx-lint version: 5.3.6
+# ------------------------------
+[MASTER]
+ignore = migrations
+persistent = yes
+load-plugins = edx_lint.pylint,pylint_django,pylint_celery
+
+[MESSAGES CONTROL]
+enable = 
+	blacklisted-name,
+	line-too-long,
+	
+	abstract-class-instantiated,
+	abstract-method,
+	access-member-before-definition,
+	anomalous-backslash-in-string,
+	anomalous-unicode-escape-in-string,
+	arguments-differ,
+	assert-on-tuple,
+	assigning-non-slot,
+	assignment-from-no-return,
+	assignment-from-none,
+	attribute-defined-outside-init,
+	bad-except-order,
+	bad-format-character,
+	bad-format-string-key,
+	bad-format-string,
+	bad-open-mode,
+	bad-reversed-sequence,
+	bad-staticmethod-argument,
+	bad-str-strip-call,
+	bad-super-call,
+	binary-op-exception,
+	boolean-datetime,
+	catching-non-exception,
+	cell-var-from-loop,
+	confusing-with-statement,
+	continue-in-finally,
+	dangerous-default-value,
+	duplicate-argument-name,
+	duplicate-bases,
+	duplicate-except,
+	duplicate-key,
+	expression-not-assigned,
+	format-combined-specification,
+	format-needs-mapping,
+	function-redefined,
+	global-variable-undefined,
+	import-error,
+	import-self,
+	inconsistent-mro,
+	inherit-non-class,
+	init-is-generator,
+	invalid-all-object,
+	invalid-format-index,
+	invalid-length-returned,
+	invalid-sequence-index,
+	invalid-slice-index,
+	invalid-slots-object,
+	invalid-slots,
+	invalid-unary-operand-type,
+	logging-too-few-args,
+	logging-too-many-args,
+	logging-unsupported-format,
+	lost-exception,
+	method-hidden,
+	misplaced-bare-raise,
+	misplaced-future,
+	missing-format-argument-key,
+	missing-format-attribute,
+	missing-format-string-key,
+	no-member,
+	no-method-argument,
+	no-name-in-module,
+	no-self-argument,
+	no-value-for-parameter,
+	non-iterator-returned,
+	non-parent-method-called,
+	nonexistent-operator,
+	not-a-mapping,
+	not-an-iterable,
+	not-callable,
+	not-context-manager,
+	not-in-loop,
+	pointless-statement,
+	pointless-string-statement,
+	raising-bad-type,
+	raising-non-exception,
+	redefined-builtin,
+	redefined-outer-name,
+	redundant-keyword-arg,
+	repeated-keyword,
+	return-arg-in-generator,
+	return-in-init,
+	return-outside-function,
+	signature-differs,
+	super-init-not-called,
+	super-method-not-called,
+	syntax-error,
+	test-inherits-tests,
+	too-few-format-args,
+	too-many-format-args,
+	too-many-function-args,
+	translation-of-non-string,
+	truncated-format-string,
+	undefined-all-variable,
+	undefined-loop-variable,
+	undefined-variable,
+	unexpected-keyword-arg,
+	unexpected-special-method-signature,
+	unpacking-non-sequence,
+	unreachable,
+	unsubscriptable-object,
+	unsupported-binary-operation,
+	unsupported-membership-test,
+	unused-format-string-argument,
+	unused-format-string-key,
+	used-before-assignment,
+	using-constant-test,
+	yield-outside-function,
+	
+	astroid-error,
+	fatal,
+	method-check-failed,
+	parse-error,
+	raw-checker-failed,
+	
+	empty-docstring,
+	invalid-characters-in-docstring,
+	missing-docstring,
+	wrong-spelling-in-comment,
+	wrong-spelling-in-docstring,
+	
+	unused-argument,
+	unused-import,
+	unused-variable,
+	
+	eval-used,
+	exec-used,
+	
+	bad-classmethod-argument,
+	bad-mcs-classmethod-argument,
+	bad-mcs-method-argument,
+	bare-except,
+	broad-except,
+	consider-iterating-dictionary,
+	consider-using-enumerate,
+	global-at-module-level,
+	global-variable-not-assigned,
+	literal-used-as-attribute,
+	logging-format-interpolation,
+	logging-not-lazy,
+	multiple-imports,
+	multiple-statements,
+	no-classmethod-decorator,
+	no-staticmethod-decorator,
+	protected-access,
+	redundant-unittest-assert,
+	reimported,
+	simplifiable-if-statement,
+	simplifiable-range,
+	singleton-comparison,
+	superfluous-parens,
+	unidiomatic-typecheck,
+	unnecessary-lambda,
+	unnecessary-pass,
+	unnecessary-semicolon,
+	unneeded-not,
+	useless-else-on-loop,
+	wrong-assert-type,
+	
+	deprecated-method,
+	deprecated-module,
+	
+	too-many-boolean-expressions,
+	too-many-nested-blocks,
+	too-many-statements,
+	
+	wildcard-import,
+	wrong-import-order,
+	wrong-import-position,
+	
+	missing-final-newline,
+	mixed-line-endings,
+	trailing-newlines,
+	trailing-whitespace,
+	unexpected-line-ending-format,
+	
+	bad-inline-option,
+	bad-option-value,
+	deprecated-pragma,
+	unrecognized-inline-option,
+	useless-suppression,
+disable = 
+	bad-indentation,
+	broad-exception-raised,
+	consider-using-f-string,
+	duplicate-code,
+	file-ignored,
+	fixme,
+	global-statement,
+	invalid-name,
+	locally-disabled,
+	no-else-return,
+	suppressed-message,
+	too-few-public-methods,
+	too-many-ancestors,
+	too-many-arguments,
+	too-many-branches,
+	too-many-instance-attributes,
+	too-many-lines,
+	too-many-locals,
+	too-many-public-methods,
+	too-many-return-statements,
+	ungrouped-imports,
+	unspecified-encoding,
+	unused-wildcard-import,
+	use-maxsplit-arg,
+	
+	feature-toggle-needs-doc,
+	illegal-waffle-usage,
+	
+	logging-fstring-interpolation,
+	django-not-configured,
+	unused-argument,
+	unsubscriptable-object
+
+[REPORTS]
+output-format = text
+reports = no
+score = no
+
+[BASIC]
+module-rgx = (([a-z_][a-z0-9_]*)|([A-Z][a-zA-Z0-9]+))$
+const-rgx = (([A-Z_][A-Z0-9_]*)|(__.*__)|log|urlpatterns)$
+class-rgx = [A-Z_][a-zA-Z0-9]+$
+function-rgx = ([a-z_][a-z0-9_]{2,40}|test_[a-z0-9_]+)$
+method-rgx = ([a-z_][a-z0-9_]{2,40}|setUp|set[Uu]pClass|tearDown|tear[Dd]ownClass|assert[A-Z]\w*|maxDiff|test_[a-z0-9_]+)$
+attr-rgx = [a-z_][a-z0-9_]{2,30}$
+argument-rgx = [a-z_][a-z0-9_]{2,30}$
+variable-rgx = [a-z_][a-z0-9_]{2,30}$
+class-attribute-rgx = ([A-Za-z_][A-Za-z0-9_]{2,30}|(__.*__))$
+inlinevar-rgx = [A-Za-z_][A-Za-z0-9_]*$
+good-names = f,i,j,k,db,ex,Run,_,__
+bad-names = foo,bar,baz,toto,tutu,tata
+no-docstring-rgx = __.*__$|test_.+|setUp$|setUpClass$|tearDown$|tearDownClass$|Meta$
+docstring-min-length = 5
+
+[FORMAT]
+max-line-length = 120
+ignore-long-lines = ^\s*(# )?((<?https?://\S+>?)|(\.\. \w+: .*))$
+single-line-if-stmt = no
+max-module-lines = 1000
+indent-string = '    '
+
+[MISCELLANEOUS]
+notes = FIXME,XXX,TODO
+
+[SIMILARITIES]
+min-similarity-lines = 4
+ignore-comments = yes
+ignore-docstrings = yes
+ignore-imports = no
+
+[TYPECHECK]
+ignore-mixin-members = yes
+ignored-classes = SQLObject
+unsafe-load-any-extension = yes
+generated-members = 
+	REQUEST,
+	acl_users,
+	aq_parent,
+	objects,
+	DoesNotExist,
+	can_read,
+	can_write,
+	get_url,
+	size,
+	content,
+	status_code,
+	create,
+	build,
+	fields,
+	tag,
+	org,
+	course,
+	category,
+	name,
+	revision,
+	_meta,
+
+[VARIABLES]
+init-import = no
+dummy-variables-rgx = _|dummy|unused|.*_unused
+additional-builtins = 
+
+[CLASSES]
+defining-attr-methods = __init__,__new__,setUp
+valid-classmethod-first-arg = cls
+valid-metaclass-classmethod-first-arg = mcs
+
+[DESIGN]
+max-args = 5
+ignored-argument-names = _.*
+max-locals = 15
+max-returns = 6
+max-branches = 12
+max-statements = 50
+max-parents = 7
+max-attributes = 7
+min-public-methods = 2
+max-public-methods = 20
+
+[IMPORTS]
+deprecated-modules = regsub,TERMIOS,Bastion,rexec
+import-graph = 
+ext-import-graph = 
+int-import-graph = 
+
+[EXCEPTIONS]
+overgeneral-exceptions = builtins.Exception
+
+# 6114ba904f03712e1def5d0f459a5ce5a0927223
diff --git a/pylintrc_tweaks b/pylintrc_tweaks
new file mode 100644
index 0000000..f24c494
--- /dev/null
+++ b/pylintrc_tweaks
@@ -0,0 +1,10 @@
+# pylintrc tweaks for use with edx_lint.
+[MASTER]
+ignore = migrations
+load-plugins = edx_lint.pylint,pylint_django,pylint_celery
+
+[MESSAGES CONTROL]
+disable+=
+    django-not-configured,
+    unused-argument,
+    unsubscriptable-object
diff --git a/requirements/base.in b/requirements/base.in
new file mode 100644
index 0000000..c02823b
--- /dev/null
+++ b/requirements/base.in
@@ -0,0 +1,6 @@
+# Core requirements for using this application
+-c constraints.txt
+
+django-statici18n
+XBlock[django]
+litellm
\ No newline at end of file
diff --git a/requirements/base.txt b/requirements/base.txt
new file mode 100644
index 0000000..6e546df
--- /dev/null
+++ b/requirements/base.txt
@@ -0,0 +1,221 @@
+#
+# This file is autogenerated by pip-compile with Python 3.8
+# by the following command:
+#
+#    make upgrade
+#
+aiohappyeyeballs==2.3.4
+    # via aiohttp
+aiohttp==3.10.0
+    # via litellm
+aiosignal==1.3.1
+    # via aiohttp
+annotated-types==0.7.0
+    # via pydantic
+anyio==4.4.0
+    # via
+    #   httpx
+    #   openai
+appdirs==1.4.4
+    # via fs
+asgiref==3.8.1
+    # via django
+async-timeout==4.0.3
+    # via aiohttp
+attrs==23.2.0
+    # via
+    #   aiohttp
+    #   jsonschema
+    #   referencing
+backports-zoneinfo==0.2.1 ; python_version < "3.9"
+    # via
+    #   -c requirements/constraints.txt
+    #   django
+boto3==1.34.151
+    # via fs-s3fs
+botocore==1.34.151
+    # via
+    #   boto3
+    #   s3transfer
+certifi==2024.7.4
+    # via
+    #   httpcore
+    #   httpx
+    #   requests
+charset-normalizer==3.3.2
+    # via requests
+click==8.1.7
+    # via litellm
+distro==1.9.0
+    # via openai
+django==4.2.14
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   django-appconf
+    #   django-statici18n
+    #   openedx-django-pyfs
+django-appconf==1.0.6
+    # via django-statici18n
+django-statici18n==2.5.0
+    # via -r requirements/base.in
+exceptiongroup==1.2.2
+    # via anyio
+filelock==3.15.4
+    # via huggingface-hub
+frozenlist==1.4.1
+    # via
+    #   aiohttp
+    #   aiosignal
+fs==2.4.16
+    # via
+    #   fs-s3fs
+    #   openedx-django-pyfs
+    #   xblock
+fs-s3fs==1.1.1
+    # via openedx-django-pyfs
+fsspec==2024.6.1
+    # via huggingface-hub
+h11==0.14.0
+    # via httpcore
+httpcore==1.0.5
+    # via httpx
+httpx==0.27.0
+    # via openai
+huggingface-hub==0.24.5
+    # via tokenizers
+idna==3.7
+    # via
+    #   anyio
+    #   httpx
+    #   requests
+    #   yarl
+importlib-metadata==6.11.0
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   litellm
+importlib-resources==6.4.0
+    # via
+    #   jsonschema
+    #   jsonschema-specifications
+jinja2==3.1.4
+    # via litellm
+jmespath==1.0.1
+    # via
+    #   boto3
+    #   botocore
+jsonschema==4.23.0
+    # via litellm
+jsonschema-specifications==2023.12.1
+    # via jsonschema
+lazy==1.6
+    # via xblock
+litellm==1.42.6
+    # via -r requirements/base.in
+lxml==5.2.2
+    # via xblock
+mako==1.3.5
+    # via xblock
+markupsafe==2.1.5
+    # via
+    #   jinja2
+    #   mako
+    #   xblock
+multidict==6.0.5
+    # via
+    #   aiohttp
+    #   yarl
+openai==1.37.1
+    # via litellm
+openedx-django-pyfs==3.6.0
+    # via xblock
+packaging==24.1
+    # via huggingface-hub
+pkgutil-resolve-name==1.3.10
+    # via jsonschema
+pydantic==2.8.2
+    # via
+    #   litellm
+    #   openai
+pydantic-core==2.20.1
+    # via pydantic
+python-dateutil==2.9.0.post0
+    # via
+    #   botocore
+    #   xblock
+python-dotenv==1.0.1
+    # via litellm
+pytz==2024.1
+    # via xblock
+pyyaml==6.0.1
+    # via
+    #   huggingface-hub
+    #   xblock
+referencing==0.35.1
+    # via
+    #   jsonschema
+    #   jsonschema-specifications
+regex==2024.7.24
+    # via tiktoken
+requests==2.32.3
+    # via
+    #   huggingface-hub
+    #   litellm
+    #   tiktoken
+rpds-py==0.19.1
+    # via
+    #   jsonschema
+    #   referencing
+s3transfer==0.10.2
+    # via boto3
+simplejson==3.19.2
+    # via xblock
+six==1.16.0
+    # via
+    #   fs
+    #   fs-s3fs
+    #   python-dateutil
+sniffio==1.3.1
+    # via
+    #   anyio
+    #   httpx
+    #   openai
+sqlparse==0.5.1
+    # via django
+tiktoken==0.7.0
+    # via litellm
+tokenizers==0.19.1
+    # via litellm
+tqdm==4.66.4
+    # via
+    #   huggingface-hub
+    #   openai
+typing-extensions==4.12.2
+    # via
+    #   annotated-types
+    #   anyio
+    #   asgiref
+    #   huggingface-hub
+    #   openai
+    #   pydantic
+    #   pydantic-core
+urllib3==1.26.19
+    # via
+    #   botocore
+    #   requests
+web-fragments==2.2.0
+    # via xblock
+webob==1.8.7
+    # via xblock
+xblock[django]==4.0.1
+    # via
+    #   -c requirements/constraints.txt
+    #   -r requirements/base.in
+yarl==1.9.4
+    # via aiohttp
+zipp==3.19.2
+    # via
+    #   importlib-metadata
+    #   importlib-resources
+
+# The following packages are considered to be unsafe in a requirements file:
+# setuptools
diff --git a/requirements/ci.in b/requirements/ci.in
new file mode 100644
index 0000000..3c1c718
--- /dev/null
+++ b/requirements/ci.in
@@ -0,0 +1,4 @@
+# Requirements for running tests in CI
+-c constraints.txt
+
+tox                       # Virtualenv management for tests
diff --git a/requirements/ci.txt b/requirements/ci.txt
new file mode 100644
index 0000000..f0993ea
--- /dev/null
+++ b/requirements/ci.txt
@@ -0,0 +1,38 @@
+#
+# This file is autogenerated by pip-compile with Python 3.8
+# by the following command:
+#
+#    make upgrade
+#
+cachetools==5.4.0
+    # via tox
+chardet==5.2.0
+    # via tox
+colorama==0.4.6
+    # via tox
+distlib==0.3.8
+    # via virtualenv
+filelock==3.15.4
+    # via
+    #   tox
+    #   virtualenv
+packaging==24.1
+    # via
+    #   pyproject-api
+    #   tox
+platformdirs==4.2.2
+    # via
+    #   tox
+    #   virtualenv
+pluggy==1.5.0
+    # via tox
+pyproject-api==1.7.1
+    # via tox
+tomli==2.0.1
+    # via
+    #   pyproject-api
+    #   tox
+tox==4.16.0
+    # via -r requirements/ci.in
+virtualenv==20.26.3
+    # via tox
diff --git a/requirements/constraints.txt b/requirements/constraints.txt
new file mode 100644
index 0000000..4146286
--- /dev/null
+++ b/requirements/constraints.txt
@@ -0,0 +1,17 @@
+# Version constraints for pip-installation.
+#
+# This file doesn't install any packages. It specifies version constraints
+# that will be applied if a package is needed.
+#
+# When pinning something here, please provide an explanation of why. Ideally,
+# link to other information that will help people in the future to remove the
+# pin when possible. Writing an issue against the offending project and
+# linking to it here is good.
+
+# Common constraints for edx repos
+-c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+
+# For python greater than or equal to 3.9 backports.zoneinfo causing failures
+backports.zoneinfo; python_version<"3.9"
+
+xblock[django]==4.0.1
\ No newline at end of file
diff --git a/requirements/dev.in b/requirements/dev.in
new file mode 100644
index 0000000..7456111
--- /dev/null
+++ b/requirements/dev.in
@@ -0,0 +1,6 @@
+# Additional requirements for development of this application
+-c constraints.txt
+
+-r pip-tools.txt          # pip-tools and its dependencies, for managing requirements files
+-r quality.txt            # Core and quality check dependencies
+-r ci.txt                 # dependencies for setting up testing in CI
diff --git a/requirements/dev.txt b/requirements/dev.txt
new file mode 100644
index 0000000..85f0896
--- /dev/null
+++ b/requirements/dev.txt
@@ -0,0 +1,579 @@
+#
+# This file is autogenerated by pip-compile with Python 3.8
+# by the following command:
+#
+#    make upgrade
+#
+aiohappyeyeballs==2.3.4
+    # via
+    #   -r requirements/quality.txt
+    #   aiohttp
+aiohttp==3.10.0
+    # via
+    #   -r requirements/quality.txt
+    #   litellm
+aiosignal==1.3.1
+    # via
+    #   -r requirements/quality.txt
+    #   aiohttp
+annotated-types==0.7.0
+    # via
+    #   -r requirements/quality.txt
+    #   pydantic
+anyio==4.4.0
+    # via
+    #   -r requirements/quality.txt
+    #   httpx
+    #   openai
+appdirs==1.4.4
+    # via
+    #   -r requirements/quality.txt
+    #   fs
+arrow==1.3.0
+    # via
+    #   -r requirements/quality.txt
+    #   cookiecutter
+asgiref==3.8.1
+    # via
+    #   -r requirements/quality.txt
+    #   django
+astroid==3.2.4
+    # via
+    #   -r requirements/quality.txt
+    #   pylint
+    #   pylint-celery
+async-timeout==4.0.3
+    # via
+    #   -r requirements/quality.txt
+    #   aiohttp
+attrs==23.2.0
+    # via
+    #   -r requirements/quality.txt
+    #   aiohttp
+    #   jsonschema
+    #   referencing
+backports-zoneinfo==0.2.1 ; python_version < "3.9"
+    # via
+    #   -c requirements/constraints.txt
+    #   -r requirements/quality.txt
+    #   django
+binaryornot==0.4.4
+    # via
+    #   -r requirements/quality.txt
+    #   cookiecutter
+boto3==1.34.151
+    # via
+    #   -r requirements/quality.txt
+    #   fs-s3fs
+botocore==1.34.151
+    # via
+    #   -r requirements/quality.txt
+    #   boto3
+    #   s3transfer
+build==1.2.1
+    # via
+    #   -r requirements/pip-tools.txt
+    #   pip-tools
+cachetools==5.4.0
+    # via
+    #   -r requirements/ci.txt
+    #   tox
+certifi==2024.7.4
+    # via
+    #   -r requirements/quality.txt
+    #   httpcore
+    #   httpx
+    #   requests
+chardet==5.2.0
+    # via
+    #   -r requirements/ci.txt
+    #   -r requirements/quality.txt
+    #   binaryornot
+    #   tox
+charset-normalizer==3.3.2
+    # via
+    #   -r requirements/quality.txt
+    #   requests
+click==8.1.7
+    # via
+    #   -r requirements/pip-tools.txt
+    #   -r requirements/quality.txt
+    #   click-log
+    #   code-annotations
+    #   cookiecutter
+    #   edx-lint
+    #   litellm
+    #   pip-tools
+click-log==0.4.0
+    # via
+    #   -r requirements/quality.txt
+    #   edx-lint
+code-annotations==1.8.0
+    # via
+    #   -r requirements/quality.txt
+    #   edx-lint
+colorama==0.4.6
+    # via
+    #   -r requirements/ci.txt
+    #   tox
+cookiecutter==2.6.0
+    # via
+    #   -r requirements/quality.txt
+    #   xblock-sdk
+coverage[toml]==7.6.0
+    # via
+    #   -r requirements/quality.txt
+    #   pytest-cov
+ddt==1.7.2
+    # via -r requirements/quality.txt
+dill==0.3.8
+    # via
+    #   -r requirements/quality.txt
+    #   pylint
+distlib==0.3.8
+    # via
+    #   -r requirements/ci.txt
+    #   virtualenv
+distro==1.9.0
+    # via
+    #   -r requirements/quality.txt
+    #   openai
+django==4.2.14
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   -r requirements/quality.txt
+    #   django-appconf
+    #   django-statici18n
+    #   edx-i18n-tools
+    #   openedx-django-pyfs
+    #   xblock-sdk
+django-appconf==1.0.6
+    # via
+    #   -r requirements/quality.txt
+    #   django-statici18n
+django-statici18n==2.5.0
+    # via -r requirements/quality.txt
+edx-i18n-tools==1.6.2
+    # via -r requirements/quality.txt
+edx-lint==5.3.7
+    # via -r requirements/quality.txt
+exceptiongroup==1.2.2
+    # via
+    #   -r requirements/quality.txt
+    #   anyio
+    #   pytest
+filelock==3.15.4
+    # via
+    #   -r requirements/ci.txt
+    #   -r requirements/quality.txt
+    #   huggingface-hub
+    #   tox
+    #   virtualenv
+frozenlist==1.4.1
+    # via
+    #   -r requirements/quality.txt
+    #   aiohttp
+    #   aiosignal
+fs==2.4.16
+    # via
+    #   -r requirements/quality.txt
+    #   fs-s3fs
+    #   openedx-django-pyfs
+    #   xblock
+fs-s3fs==1.1.1
+    # via
+    #   -r requirements/quality.txt
+    #   openedx-django-pyfs
+    #   xblock-sdk
+fsspec==2024.6.1
+    # via
+    #   -r requirements/quality.txt
+    #   huggingface-hub
+h11==0.14.0
+    # via
+    #   -r requirements/quality.txt
+    #   httpcore
+httpcore==1.0.5
+    # via
+    #   -r requirements/quality.txt
+    #   httpx
+httpx==0.27.0
+    # via
+    #   -r requirements/quality.txt
+    #   openai
+huggingface-hub==0.24.5
+    # via
+    #   -r requirements/quality.txt
+    #   tokenizers
+idna==3.7
+    # via
+    #   -r requirements/quality.txt
+    #   anyio
+    #   httpx
+    #   requests
+    #   yarl
+importlib-metadata==6.11.0
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   -r requirements/pip-tools.txt
+    #   -r requirements/quality.txt
+    #   build
+    #   litellm
+importlib-resources==6.4.0
+    # via
+    #   -r requirements/quality.txt
+    #   jsonschema
+    #   jsonschema-specifications
+iniconfig==2.0.0
+    # via
+    #   -r requirements/quality.txt
+    #   pytest
+isort==5.13.2
+    # via
+    #   -r requirements/quality.txt
+    #   pylint
+jinja2==3.1.4
+    # via
+    #   -r requirements/quality.txt
+    #   code-annotations
+    #   cookiecutter
+    #   litellm
+jmespath==1.0.1
+    # via
+    #   -r requirements/quality.txt
+    #   boto3
+    #   botocore
+jsonschema==4.23.0
+    # via
+    #   -r requirements/quality.txt
+    #   litellm
+jsonschema-specifications==2023.12.1
+    # via
+    #   -r requirements/quality.txt
+    #   jsonschema
+lazy==1.6
+    # via
+    #   -r requirements/quality.txt
+    #   xblock
+litellm==1.42.6
+    # via -r requirements/quality.txt
+lxml==5.2.2
+    # via
+    #   -r requirements/quality.txt
+    #   edx-i18n-tools
+    #   xblock
+    #   xblock-sdk
+mako==1.3.5
+    # via
+    #   -r requirements/quality.txt
+    #   xblock
+markdown-it-py==3.0.0
+    # via
+    #   -r requirements/quality.txt
+    #   rich
+markupsafe==2.1.5
+    # via
+    #   -r requirements/quality.txt
+    #   jinja2
+    #   mako
+    #   xblock
+mccabe==0.7.0
+    # via
+    #   -r requirements/quality.txt
+    #   pylint
+mdurl==0.1.2
+    # via
+    #   -r requirements/quality.txt
+    #   markdown-it-py
+mock==5.1.0
+    # via -r requirements/quality.txt
+multidict==6.0.5
+    # via
+    #   -r requirements/quality.txt
+    #   aiohttp
+    #   yarl
+openai==1.37.1
+    # via
+    #   -r requirements/quality.txt
+    #   litellm
+openedx-django-pyfs==3.6.0
+    # via
+    #   -r requirements/quality.txt
+    #   xblock
+packaging==24.1
+    # via
+    #   -r requirements/ci.txt
+    #   -r requirements/pip-tools.txt
+    #   -r requirements/quality.txt
+    #   build
+    #   huggingface-hub
+    #   pyproject-api
+    #   pytest
+    #   tox
+path==16.16.0
+    # via
+    #   -r requirements/quality.txt
+    #   edx-i18n-tools
+pbr==6.0.0
+    # via
+    #   -r requirements/quality.txt
+    #   stevedore
+pip-tools==7.4.1
+    # via -r requirements/pip-tools.txt
+pkgutil-resolve-name==1.3.10
+    # via
+    #   -r requirements/quality.txt
+    #   jsonschema
+platformdirs==4.2.2
+    # via
+    #   -r requirements/ci.txt
+    #   -r requirements/quality.txt
+    #   pylint
+    #   tox
+    #   virtualenv
+pluggy==1.5.0
+    # via
+    #   -r requirements/ci.txt
+    #   -r requirements/quality.txt
+    #   pytest
+    #   tox
+polib==1.2.0
+    # via
+    #   -r requirements/quality.txt
+    #   edx-i18n-tools
+pycodestyle==2.12.0
+    # via -r requirements/quality.txt
+pydantic==2.8.2
+    # via
+    #   -r requirements/quality.txt
+    #   litellm
+    #   openai
+pydantic-core==2.20.1
+    # via
+    #   -r requirements/quality.txt
+    #   pydantic
+pygments==2.18.0
+    # via
+    #   -r requirements/quality.txt
+    #   rich
+pylint==3.2.6
+    # via
+    #   -r requirements/quality.txt
+    #   edx-lint
+    #   pylint-celery
+    #   pylint-django
+    #   pylint-plugin-utils
+pylint-celery==0.3
+    # via
+    #   -r requirements/quality.txt
+    #   edx-lint
+pylint-django==2.5.5
+    # via
+    #   -r requirements/quality.txt
+    #   edx-lint
+pylint-plugin-utils==0.8.2
+    # via
+    #   -r requirements/quality.txt
+    #   pylint-celery
+    #   pylint-django
+pypng==0.20220715.0
+    # via
+    #   -r requirements/quality.txt
+    #   xblock-sdk
+pyproject-api==1.7.1
+    # via
+    #   -r requirements/ci.txt
+    #   tox
+pyproject-hooks==1.1.0
+    # via
+    #   -r requirements/pip-tools.txt
+    #   build
+    #   pip-tools
+pytest==8.3.2
+    # via
+    #   -r requirements/quality.txt
+    #   pytest-cov
+    #   pytest-django
+pytest-cov==5.0.0
+    # via -r requirements/quality.txt
+pytest-django==4.8.0
+    # via -r requirements/quality.txt
+python-dateutil==2.9.0.post0
+    # via
+    #   -r requirements/quality.txt
+    #   arrow
+    #   botocore
+    #   xblock
+python-dotenv==1.0.1
+    # via
+    #   -r requirements/quality.txt
+    #   litellm
+python-slugify==8.0.4
+    # via
+    #   -r requirements/quality.txt
+    #   code-annotations
+    #   cookiecutter
+pytz==2024.1
+    # via
+    #   -r requirements/quality.txt
+    #   xblock
+pyyaml==6.0.1
+    # via
+    #   -r requirements/quality.txt
+    #   code-annotations
+    #   cookiecutter
+    #   edx-i18n-tools
+    #   huggingface-hub
+    #   xblock
+referencing==0.35.1
+    # via
+    #   -r requirements/quality.txt
+    #   jsonschema
+    #   jsonschema-specifications
+regex==2024.7.24
+    # via
+    #   -r requirements/quality.txt
+    #   tiktoken
+requests==2.32.3
+    # via
+    #   -r requirements/quality.txt
+    #   cookiecutter
+    #   huggingface-hub
+    #   litellm
+    #   tiktoken
+    #   xblock-sdk
+rich==13.7.1
+    # via
+    #   -r requirements/quality.txt
+    #   cookiecutter
+rpds-py==0.19.1
+    # via
+    #   -r requirements/quality.txt
+    #   jsonschema
+    #   referencing
+s3transfer==0.10.2
+    # via
+    #   -r requirements/quality.txt
+    #   boto3
+simplejson==3.19.2
+    # via
+    #   -r requirements/quality.txt
+    #   xblock
+    #   xblock-sdk
+six==1.16.0
+    # via
+    #   -r requirements/quality.txt
+    #   edx-lint
+    #   fs
+    #   fs-s3fs
+    #   python-dateutil
+sniffio==1.3.1
+    # via
+    #   -r requirements/quality.txt
+    #   anyio
+    #   httpx
+    #   openai
+sqlparse==0.5.1
+    # via
+    #   -r requirements/quality.txt
+    #   django
+stevedore==5.2.0
+    # via
+    #   -r requirements/quality.txt
+    #   code-annotations
+text-unidecode==1.3
+    # via
+    #   -r requirements/quality.txt
+    #   python-slugify
+tiktoken==0.7.0
+    # via
+    #   -r requirements/quality.txt
+    #   litellm
+tokenizers==0.19.1
+    # via
+    #   -r requirements/quality.txt
+    #   litellm
+tomli==2.0.1
+    # via
+    #   -r requirements/ci.txt
+    #   -r requirements/pip-tools.txt
+    #   -r requirements/quality.txt
+    #   build
+    #   coverage
+    #   pip-tools
+    #   pylint
+    #   pyproject-api
+    #   pytest
+    #   tox
+tomlkit==0.13.0
+    # via
+    #   -r requirements/quality.txt
+    #   pylint
+tox==4.16.0
+    # via -r requirements/ci.txt
+tqdm==4.66.4
+    # via
+    #   -r requirements/quality.txt
+    #   huggingface-hub
+    #   openai
+types-python-dateutil==2.9.0.20240316
+    # via
+    #   -r requirements/quality.txt
+    #   arrow
+typing-extensions==4.12.2
+    # via
+    #   -r requirements/quality.txt
+    #   annotated-types
+    #   anyio
+    #   asgiref
+    #   astroid
+    #   huggingface-hub
+    #   openai
+    #   pydantic
+    #   pydantic-core
+    #   pylint
+    #   rich
+urllib3==1.26.19
+    # via
+    #   -r requirements/quality.txt
+    #   botocore
+    #   requests
+virtualenv==20.26.3
+    # via
+    #   -r requirements/ci.txt
+    #   tox
+web-fragments==2.2.0
+    # via
+    #   -r requirements/quality.txt
+    #   xblock
+    #   xblock-sdk
+webob==1.8.7
+    # via
+    #   -r requirements/quality.txt
+    #   xblock
+    #   xblock-sdk
+wheel==0.43.0
+    # via
+    #   -r requirements/pip-tools.txt
+    #   pip-tools
+xblock[django]==4.0.1
+    # via
+    #   -c requirements/constraints.txt
+    #   -r requirements/quality.txt
+    #   xblock-sdk
+xblock-sdk==0.11.0
+    # via -r requirements/quality.txt
+yarl==1.9.4
+    # via
+    #   -r requirements/quality.txt
+    #   aiohttp
+zipp==3.19.2
+    # via
+    #   -r requirements/pip-tools.txt
+    #   -r requirements/quality.txt
+    #   importlib-metadata
+    #   importlib-resources
+
+# The following packages are considered to be unsafe in a requirements file:
+# pip
+# setuptools
diff --git a/requirements/pip-tools.in b/requirements/pip-tools.in
new file mode 100644
index 0000000..0295d2c
--- /dev/null
+++ b/requirements/pip-tools.in
@@ -0,0 +1,5 @@
+# Just the dependencies to run pip-tools, mainly for the "upgrade" make target
+
+-c constraints.txt
+
+pip-tools                 # Contains pip-compile, used to generate pip requirements files
diff --git a/requirements/pip-tools.txt b/requirements/pip-tools.txt
new file mode 100644
index 0000000..3058830
--- /dev/null
+++ b/requirements/pip-tools.txt
@@ -0,0 +1,34 @@
+#
+# This file is autogenerated by pip-compile with Python 3.8
+# by the following command:
+#
+#    make upgrade
+#
+build==1.2.1
+    # via pip-tools
+click==8.1.7
+    # via pip-tools
+importlib-metadata==6.11.0
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   build
+packaging==24.1
+    # via build
+pip-tools==7.4.1
+    # via -r requirements/pip-tools.in
+pyproject-hooks==1.1.0
+    # via
+    #   build
+    #   pip-tools
+tomli==2.0.1
+    # via
+    #   build
+    #   pip-tools
+wheel==0.43.0
+    # via pip-tools
+zipp==3.19.2
+    # via importlib-metadata
+
+# The following packages are considered to be unsafe in a requirements file:
+# pip
+# setuptools
diff --git a/requirements/pip.in b/requirements/pip.in
new file mode 100644
index 0000000..716c6f2
--- /dev/null
+++ b/requirements/pip.in
@@ -0,0 +1,6 @@
+# Core dependencies for installing other packages
+-c constraints.txt
+
+pip
+setuptools
+wheel
diff --git a/requirements/pip.txt b/requirements/pip.txt
new file mode 100644
index 0000000..7b187b8
--- /dev/null
+++ b/requirements/pip.txt
@@ -0,0 +1,14 @@
+#
+# This file is autogenerated by pip-compile with Python 3.8
+# by the following command:
+#
+#    make upgrade
+#
+wheel==0.43.0
+    # via -r requirements/pip.in
+
+# The following packages are considered to be unsafe in a requirements file:
+pip==24.2
+    # via -r requirements/pip.in
+setuptools==72.1.0
+    # via -r requirements/pip.in
diff --git a/requirements/private.readme b/requirements/private.readme
new file mode 100644
index 0000000..5600a10
--- /dev/null
+++ b/requirements/private.readme
@@ -0,0 +1,15 @@
+# If there are any Python packages you want to keep in your virtualenv beyond
+# those listed in the official requirements files, create a "private.in" file
+# and list them there.  Generate the corresponding "private.txt" file pinning
+# all of their indirect dependencies to specific versions as follows:
+
+#     pip-compile private.in
+
+# This allows you to use "pip-sync" without removing these packages:
+
+#     pip-sync requirements/*.txt
+
+# "private.in" and "private.txt" aren't checked into git to avoid merge
+# conflicts, and the presence of this file allows "private.*" to be
+# included in scripted pip-sync usage without requiring that those files be
+# created first.
diff --git a/requirements/quality.in b/requirements/quality.in
new file mode 100644
index 0000000..76e2b39
--- /dev/null
+++ b/requirements/quality.in
@@ -0,0 +1,7 @@
+# Requirements for code quality checks
+-c constraints.txt
+
+-r test.txt               # Core and testing dependencies for this package
+
+edx-lint                  # edX pylint rules and plugins
+pycodestyle               # PEP 8 compliance validation
diff --git a/requirements/quality.txt b/requirements/quality.txt
new file mode 100644
index 0000000..a8f70cb
--- /dev/null
+++ b/requirements/quality.txt
@@ -0,0 +1,490 @@
+#
+# This file is autogenerated by pip-compile with Python 3.8
+# by the following command:
+#
+#    make upgrade
+#
+aiohappyeyeballs==2.3.4
+    # via
+    #   -r requirements/test.txt
+    #   aiohttp
+aiohttp==3.10.0
+    # via
+    #   -r requirements/test.txt
+    #   litellm
+aiosignal==1.3.1
+    # via
+    #   -r requirements/test.txt
+    #   aiohttp
+annotated-types==0.7.0
+    # via
+    #   -r requirements/test.txt
+    #   pydantic
+anyio==4.4.0
+    # via
+    #   -r requirements/test.txt
+    #   httpx
+    #   openai
+appdirs==1.4.4
+    # via
+    #   -r requirements/test.txt
+    #   fs
+arrow==1.3.0
+    # via
+    #   -r requirements/test.txt
+    #   cookiecutter
+asgiref==3.8.1
+    # via
+    #   -r requirements/test.txt
+    #   django
+astroid==3.2.4
+    # via
+    #   pylint
+    #   pylint-celery
+async-timeout==4.0.3
+    # via
+    #   -r requirements/test.txt
+    #   aiohttp
+attrs==23.2.0
+    # via
+    #   -r requirements/test.txt
+    #   aiohttp
+    #   jsonschema
+    #   referencing
+backports-zoneinfo==0.2.1 ; python_version < "3.9"
+    # via
+    #   -c requirements/constraints.txt
+    #   -r requirements/test.txt
+    #   django
+binaryornot==0.4.4
+    # via
+    #   -r requirements/test.txt
+    #   cookiecutter
+boto3==1.34.151
+    # via
+    #   -r requirements/test.txt
+    #   fs-s3fs
+botocore==1.34.151
+    # via
+    #   -r requirements/test.txt
+    #   boto3
+    #   s3transfer
+certifi==2024.7.4
+    # via
+    #   -r requirements/test.txt
+    #   httpcore
+    #   httpx
+    #   requests
+chardet==5.2.0
+    # via
+    #   -r requirements/test.txt
+    #   binaryornot
+charset-normalizer==3.3.2
+    # via
+    #   -r requirements/test.txt
+    #   requests
+click==8.1.7
+    # via
+    #   -r requirements/test.txt
+    #   click-log
+    #   code-annotations
+    #   cookiecutter
+    #   edx-lint
+    #   litellm
+click-log==0.4.0
+    # via edx-lint
+code-annotations==1.8.0
+    # via edx-lint
+cookiecutter==2.6.0
+    # via
+    #   -r requirements/test.txt
+    #   xblock-sdk
+coverage[toml]==7.6.0
+    # via
+    #   -r requirements/test.txt
+    #   pytest-cov
+ddt==1.7.2
+    # via -r requirements/test.txt
+dill==0.3.8
+    # via pylint
+distro==1.9.0
+    # via
+    #   -r requirements/test.txt
+    #   openai
+django==4.2.14
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   -r requirements/test.txt
+    #   django-appconf
+    #   django-statici18n
+    #   edx-i18n-tools
+    #   openedx-django-pyfs
+    #   xblock-sdk
+django-appconf==1.0.6
+    # via
+    #   -r requirements/test.txt
+    #   django-statici18n
+django-statici18n==2.5.0
+    # via -r requirements/test.txt
+edx-i18n-tools==1.6.2
+    # via -r requirements/test.txt
+edx-lint==5.3.7
+    # via -r requirements/quality.in
+exceptiongroup==1.2.2
+    # via
+    #   -r requirements/test.txt
+    #   anyio
+    #   pytest
+filelock==3.15.4
+    # via
+    #   -r requirements/test.txt
+    #   huggingface-hub
+frozenlist==1.4.1
+    # via
+    #   -r requirements/test.txt
+    #   aiohttp
+    #   aiosignal
+fs==2.4.16
+    # via
+    #   -r requirements/test.txt
+    #   fs-s3fs
+    #   openedx-django-pyfs
+    #   xblock
+fs-s3fs==1.1.1
+    # via
+    #   -r requirements/test.txt
+    #   openedx-django-pyfs
+    #   xblock-sdk
+fsspec==2024.6.1
+    # via
+    #   -r requirements/test.txt
+    #   huggingface-hub
+h11==0.14.0
+    # via
+    #   -r requirements/test.txt
+    #   httpcore
+httpcore==1.0.5
+    # via
+    #   -r requirements/test.txt
+    #   httpx
+httpx==0.27.0
+    # via
+    #   -r requirements/test.txt
+    #   openai
+huggingface-hub==0.24.5
+    # via
+    #   -r requirements/test.txt
+    #   tokenizers
+idna==3.7
+    # via
+    #   -r requirements/test.txt
+    #   anyio
+    #   httpx
+    #   requests
+    #   yarl
+importlib-metadata==6.11.0
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   -r requirements/test.txt
+    #   litellm
+importlib-resources==6.4.0
+    # via
+    #   -r requirements/test.txt
+    #   jsonschema
+    #   jsonschema-specifications
+iniconfig==2.0.0
+    # via
+    #   -r requirements/test.txt
+    #   pytest
+isort==5.13.2
+    # via pylint
+jinja2==3.1.4
+    # via
+    #   -r requirements/test.txt
+    #   code-annotations
+    #   cookiecutter
+    #   litellm
+jmespath==1.0.1
+    # via
+    #   -r requirements/test.txt
+    #   boto3
+    #   botocore
+jsonschema==4.23.0
+    # via
+    #   -r requirements/test.txt
+    #   litellm
+jsonschema-specifications==2023.12.1
+    # via
+    #   -r requirements/test.txt
+    #   jsonschema
+lazy==1.6
+    # via
+    #   -r requirements/test.txt
+    #   xblock
+litellm==1.42.6
+    # via -r requirements/test.txt
+lxml==5.2.2
+    # via
+    #   -r requirements/test.txt
+    #   edx-i18n-tools
+    #   xblock
+    #   xblock-sdk
+mako==1.3.5
+    # via
+    #   -r requirements/test.txt
+    #   xblock
+markdown-it-py==3.0.0
+    # via
+    #   -r requirements/test.txt
+    #   rich
+markupsafe==2.1.5
+    # via
+    #   -r requirements/test.txt
+    #   jinja2
+    #   mako
+    #   xblock
+mccabe==0.7.0
+    # via pylint
+mdurl==0.1.2
+    # via
+    #   -r requirements/test.txt
+    #   markdown-it-py
+mock==5.1.0
+    # via -r requirements/test.txt
+multidict==6.0.5
+    # via
+    #   -r requirements/test.txt
+    #   aiohttp
+    #   yarl
+openai==1.37.1
+    # via
+    #   -r requirements/test.txt
+    #   litellm
+openedx-django-pyfs==3.6.0
+    # via
+    #   -r requirements/test.txt
+    #   xblock
+packaging==24.1
+    # via
+    #   -r requirements/test.txt
+    #   huggingface-hub
+    #   pytest
+path==16.16.0
+    # via
+    #   -r requirements/test.txt
+    #   edx-i18n-tools
+pbr==6.0.0
+    # via stevedore
+pkgutil-resolve-name==1.3.10
+    # via
+    #   -r requirements/test.txt
+    #   jsonschema
+platformdirs==4.2.2
+    # via pylint
+pluggy==1.5.0
+    # via
+    #   -r requirements/test.txt
+    #   pytest
+polib==1.2.0
+    # via
+    #   -r requirements/test.txt
+    #   edx-i18n-tools
+pycodestyle==2.12.0
+    # via -r requirements/quality.in
+pydantic==2.8.2
+    # via
+    #   -r requirements/test.txt
+    #   litellm
+    #   openai
+pydantic-core==2.20.1
+    # via
+    #   -r requirements/test.txt
+    #   pydantic
+pygments==2.18.0
+    # via
+    #   -r requirements/test.txt
+    #   rich
+pylint==3.2.6
+    # via
+    #   edx-lint
+    #   pylint-celery
+    #   pylint-django
+    #   pylint-plugin-utils
+pylint-celery==0.3
+    # via edx-lint
+pylint-django==2.5.5
+    # via edx-lint
+pylint-plugin-utils==0.8.2
+    # via
+    #   pylint-celery
+    #   pylint-django
+pypng==0.20220715.0
+    # via
+    #   -r requirements/test.txt
+    #   xblock-sdk
+pytest==8.3.2
+    # via
+    #   -r requirements/test.txt
+    #   pytest-cov
+    #   pytest-django
+pytest-cov==5.0.0
+    # via -r requirements/test.txt
+pytest-django==4.8.0
+    # via -r requirements/test.txt
+python-dateutil==2.9.0.post0
+    # via
+    #   -r requirements/test.txt
+    #   arrow
+    #   botocore
+    #   xblock
+python-dotenv==1.0.1
+    # via
+    #   -r requirements/test.txt
+    #   litellm
+python-slugify==8.0.4
+    # via
+    #   -r requirements/test.txt
+    #   code-annotations
+    #   cookiecutter
+pytz==2024.1
+    # via
+    #   -r requirements/test.txt
+    #   xblock
+pyyaml==6.0.1
+    # via
+    #   -r requirements/test.txt
+    #   code-annotations
+    #   cookiecutter
+    #   edx-i18n-tools
+    #   huggingface-hub
+    #   xblock
+referencing==0.35.1
+    # via
+    #   -r requirements/test.txt
+    #   jsonschema
+    #   jsonschema-specifications
+regex==2024.7.24
+    # via
+    #   -r requirements/test.txt
+    #   tiktoken
+requests==2.32.3
+    # via
+    #   -r requirements/test.txt
+    #   cookiecutter
+    #   huggingface-hub
+    #   litellm
+    #   tiktoken
+    #   xblock-sdk
+rich==13.7.1
+    # via
+    #   -r requirements/test.txt
+    #   cookiecutter
+rpds-py==0.19.1
+    # via
+    #   -r requirements/test.txt
+    #   jsonschema
+    #   referencing
+s3transfer==0.10.2
+    # via
+    #   -r requirements/test.txt
+    #   boto3
+simplejson==3.19.2
+    # via
+    #   -r requirements/test.txt
+    #   xblock
+    #   xblock-sdk
+six==1.16.0
+    # via
+    #   -r requirements/test.txt
+    #   edx-lint
+    #   fs
+    #   fs-s3fs
+    #   python-dateutil
+sniffio==1.3.1
+    # via
+    #   -r requirements/test.txt
+    #   anyio
+    #   httpx
+    #   openai
+sqlparse==0.5.1
+    # via
+    #   -r requirements/test.txt
+    #   django
+stevedore==5.2.0
+    # via code-annotations
+text-unidecode==1.3
+    # via
+    #   -r requirements/test.txt
+    #   python-slugify
+tiktoken==0.7.0
+    # via
+    #   -r requirements/test.txt
+    #   litellm
+tokenizers==0.19.1
+    # via
+    #   -r requirements/test.txt
+    #   litellm
+tomli==2.0.1
+    # via
+    #   -r requirements/test.txt
+    #   coverage
+    #   pylint
+    #   pytest
+tomlkit==0.13.0
+    # via pylint
+tqdm==4.66.4
+    # via
+    #   -r requirements/test.txt
+    #   huggingface-hub
+    #   openai
+types-python-dateutil==2.9.0.20240316
+    # via
+    #   -r requirements/test.txt
+    #   arrow
+typing-extensions==4.12.2
+    # via
+    #   -r requirements/test.txt
+    #   annotated-types
+    #   anyio
+    #   asgiref
+    #   astroid
+    #   huggingface-hub
+    #   openai
+    #   pydantic
+    #   pydantic-core
+    #   pylint
+    #   rich
+urllib3==1.26.19
+    # via
+    #   -r requirements/test.txt
+    #   botocore
+    #   requests
+web-fragments==2.2.0
+    # via
+    #   -r requirements/test.txt
+    #   xblock
+    #   xblock-sdk
+webob==1.8.7
+    # via
+    #   -r requirements/test.txt
+    #   xblock
+    #   xblock-sdk
+xblock[django]==4.0.1
+    # via
+    #   -c requirements/constraints.txt
+    #   -r requirements/test.txt
+    #   xblock-sdk
+xblock-sdk==0.11.0
+    # via -r requirements/test.txt
+yarl==1.9.4
+    # via
+    #   -r requirements/test.txt
+    #   aiohttp
+zipp==3.19.2
+    # via
+    #   -r requirements/test.txt
+    #   importlib-metadata
+    #   importlib-resources
+
+# The following packages are considered to be unsafe in a requirements file:
+# setuptools
diff --git a/requirements/test.in b/requirements/test.in
new file mode 100644
index 0000000..d5bb585
--- /dev/null
+++ b/requirements/test.in
@@ -0,0 +1,15 @@
+# Requirements for test runs.
+-c constraints.txt
+
+-r base.txt               # Core dependencies for this package
+
+pytest-cov                # pytest extension for code coverage statistics
+pytest-django             # pytest extension for better Django support
+ddt                       # data-driven tests
+
+mock                      # required by the workbench
+openedx-django-pyfs       # required by the workbench
+
+edx-i18n-tools            # For i18n_tool dummy
+
+xblock-sdk>0.7            # workbench
diff --git a/requirements/test.txt b/requirements/test.txt
new file mode 100644
index 0000000..39a089c
--- /dev/null
+++ b/requirements/test.txt
@@ -0,0 +1,404 @@
+#
+# This file is autogenerated by pip-compile with Python 3.8
+# by the following command:
+#
+#    make upgrade
+#
+aiohappyeyeballs==2.3.4
+    # via
+    #   -r requirements/base.txt
+    #   aiohttp
+aiohttp==3.10.0
+    # via
+    #   -r requirements/base.txt
+    #   litellm
+aiosignal==1.3.1
+    # via
+    #   -r requirements/base.txt
+    #   aiohttp
+annotated-types==0.7.0
+    # via
+    #   -r requirements/base.txt
+    #   pydantic
+anyio==4.4.0
+    # via
+    #   -r requirements/base.txt
+    #   httpx
+    #   openai
+appdirs==1.4.4
+    # via
+    #   -r requirements/base.txt
+    #   fs
+arrow==1.3.0
+    # via cookiecutter
+asgiref==3.8.1
+    # via
+    #   -r requirements/base.txt
+    #   django
+async-timeout==4.0.3
+    # via
+    #   -r requirements/base.txt
+    #   aiohttp
+attrs==23.2.0
+    # via
+    #   -r requirements/base.txt
+    #   aiohttp
+    #   jsonschema
+    #   referencing
+backports-zoneinfo==0.2.1 ; python_version < "3.9"
+    # via
+    #   -c requirements/constraints.txt
+    #   -r requirements/base.txt
+    #   django
+binaryornot==0.4.4
+    # via cookiecutter
+boto3==1.34.151
+    # via
+    #   -r requirements/base.txt
+    #   fs-s3fs
+botocore==1.34.151
+    # via
+    #   -r requirements/base.txt
+    #   boto3
+    #   s3transfer
+certifi==2024.7.4
+    # via
+    #   -r requirements/base.txt
+    #   httpcore
+    #   httpx
+    #   requests
+chardet==5.2.0
+    # via binaryornot
+charset-normalizer==3.3.2
+    # via
+    #   -r requirements/base.txt
+    #   requests
+click==8.1.7
+    # via
+    #   -r requirements/base.txt
+    #   cookiecutter
+    #   litellm
+cookiecutter==2.6.0
+    # via xblock-sdk
+coverage[toml]==7.6.0
+    # via pytest-cov
+ddt==1.7.2
+    # via -r requirements/test.in
+distro==1.9.0
+    # via
+    #   -r requirements/base.txt
+    #   openai
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   -r requirements/base.txt
+    #   django-appconf
+    #   django-statici18n
+    #   edx-i18n-tools
+    #   openedx-django-pyfs
+    #   xblock-sdk
+django-appconf==1.0.6
+    # via
+    #   -r requirements/base.txt
+    #   django-statici18n
+django-statici18n==2.5.0
+    # via -r requirements/base.txt
+edx-i18n-tools==1.6.2
+    # via -r requirements/test.in
+exceptiongroup==1.2.2
+    # via
+    #   -r requirements/base.txt
+    #   anyio
+    #   pytest
+filelock==3.15.4
+    # via
+    #   -r requirements/base.txt
+    #   huggingface-hub
+frozenlist==1.4.1
+    # via
+    #   -r requirements/base.txt
+    #   aiohttp
+    #   aiosignal
+fs==2.4.16
+    # via
+    #   -r requirements/base.txt
+    #   fs-s3fs
+    #   openedx-django-pyfs
+    #   xblock
+fs-s3fs==1.1.1
+    # via
+    #   -r requirements/base.txt
+    #   openedx-django-pyfs
+    #   xblock-sdk
+fsspec==2024.6.1
+    # via
+    #   -r requirements/base.txt
+    #   huggingface-hub
+h11==0.14.0
+    # via
+    #   -r requirements/base.txt
+    #   httpcore
+httpcore==1.0.5
+    # via
+    #   -r requirements/base.txt
+    #   httpx
+httpx==0.27.0
+    # via
+    #   -r requirements/base.txt
+    #   openai
+huggingface-hub==0.24.5
+    # via
+    #   -r requirements/base.txt
+    #   tokenizers
+idna==3.7
+    # via
+    #   -r requirements/base.txt
+    #   anyio
+    #   httpx
+    #   requests
+    #   yarl
+importlib-metadata==6.11.0
+    # via
+    #   -c https://raw.githubusercontent.com/edx/edx-lint/master/edx_lint/files/common_constraints.txt
+    #   -r requirements/base.txt
+    #   litellm
+importlib-resources==6.4.0
+    # via
+    #   -r requirements/base.txt
+    #   jsonschema
+    #   jsonschema-specifications
+iniconfig==2.0.0
+    # via pytest
+jinja2==3.1.4
+    # via
+    #   -r requirements/base.txt
+    #   cookiecutter
+    #   litellm
+jmespath==1.0.1
+    # via
+    #   -r requirements/base.txt
+    #   boto3
+    #   botocore
+jsonschema==4.23.0
+    # via
+    #   -r requirements/base.txt
+    #   litellm
+jsonschema-specifications==2023.12.1
+    # via
+    #   -r requirements/base.txt
+    #   jsonschema
+lazy==1.6
+    # via
+    #   -r requirements/base.txt
+    #   xblock
+litellm==1.42.6
+    # via -r requirements/base.txt
+lxml==5.2.2
+    # via
+    #   -r requirements/base.txt
+    #   edx-i18n-tools
+    #   xblock
+    #   xblock-sdk
+mako==1.3.5
+    # via
+    #   -r requirements/base.txt
+    #   xblock
+markdown-it-py==3.0.0
+    # via rich
+markupsafe==2.1.5
+    # via
+    #   -r requirements/base.txt
+    #   jinja2
+    #   mako
+    #   xblock
+mdurl==0.1.2
+    # via markdown-it-py
+mock==5.1.0
+    # via -r requirements/test.in
+multidict==6.0.5
+    # via
+    #   -r requirements/base.txt
+    #   aiohttp
+    #   yarl
+openai==1.37.1
+    # via
+    #   -r requirements/base.txt
+    #   litellm
+openedx-django-pyfs==3.6.0
+    # via
+    #   -r requirements/base.txt
+    #   -r requirements/test.in
+    #   xblock
+packaging==24.1
+    # via
+    #   -r requirements/base.txt
+    #   huggingface-hub
+    #   pytest
+path==16.16.0
+    # via edx-i18n-tools
+pkgutil-resolve-name==1.3.10
+    # via
+    #   -r requirements/base.txt
+    #   jsonschema
+pluggy==1.5.0
+    # via pytest
+polib==1.2.0
+    # via edx-i18n-tools
+pydantic==2.8.2
+    # via
+    #   -r requirements/base.txt
+    #   litellm
+    #   openai
+pydantic-core==2.20.1
+    # via
+    #   -r requirements/base.txt
+    #   pydantic
+pygments==2.18.0
+    # via rich
+pypng==0.20220715.0
+    # via xblock-sdk
+pytest==8.3.2
+    # via
+    #   pytest-cov
+    #   pytest-django
+pytest-cov==5.0.0
+    # via -r requirements/test.in
+pytest-django==4.8.0
+    # via -r requirements/test.in
+python-dateutil==2.9.0.post0
+    # via
+    #   -r requirements/base.txt
+    #   arrow
+    #   botocore
+    #   xblock
+python-dotenv==1.0.1
+    # via
+    #   -r requirements/base.txt
+    #   litellm
+python-slugify==8.0.4
+    # via cookiecutter
+pytz==2024.1
+    # via
+    #   -r requirements/base.txt
+    #   xblock
+pyyaml==6.0.1
+    # via
+    #   -r requirements/base.txt
+    #   cookiecutter
+    #   edx-i18n-tools
+    #   huggingface-hub
+    #   xblock
+referencing==0.35.1
+    # via
+    #   -r requirements/base.txt
+    #   jsonschema
+    #   jsonschema-specifications
+regex==2024.7.24
+    # via
+    #   -r requirements/base.txt
+    #   tiktoken
+requests==2.32.3
+    # via
+    #   -r requirements/base.txt
+    #   cookiecutter
+    #   huggingface-hub
+    #   litellm
+    #   tiktoken
+    #   xblock-sdk
+rich==13.7.1
+    # via cookiecutter
+rpds-py==0.19.1
+    # via
+    #   -r requirements/base.txt
+    #   jsonschema
+    #   referencing
+s3transfer==0.10.2
+    # via
+    #   -r requirements/base.txt
+    #   boto3
+simplejson==3.19.2
+    # via
+    #   -r requirements/base.txt
+    #   xblock
+    #   xblock-sdk
+six==1.16.0
+    # via
+    #   -r requirements/base.txt
+    #   fs
+    #   fs-s3fs
+    #   python-dateutil
+sniffio==1.3.1
+    # via
+    #   -r requirements/base.txt
+    #   anyio
+    #   httpx
+    #   openai
+sqlparse==0.5.1
+    # via
+    #   -r requirements/base.txt
+    #   django
+text-unidecode==1.3
+    # via python-slugify
+tiktoken==0.7.0
+    # via
+    #   -r requirements/base.txt
+    #   litellm
+tokenizers==0.19.1
+    # via
+    #   -r requirements/base.txt
+    #   litellm
+tomli==2.0.1
+    # via
+    #   coverage
+    #   pytest
+tqdm==4.66.4
+    # via
+    #   -r requirements/base.txt
+    #   huggingface-hub
+    #   openai
+types-python-dateutil==2.9.0.20240316
+    # via arrow
+typing-extensions==4.12.2
+    # via
+    #   -r requirements/base.txt
+    #   annotated-types
+    #   anyio
+    #   asgiref
+    #   huggingface-hub
+    #   openai
+    #   pydantic
+    #   pydantic-core
+    #   rich
+urllib3==1.26.19
+    # via
+    #   -r requirements/base.txt
+    #   botocore
+    #   requests
+web-fragments==2.2.0
+    # via
+    #   -r requirements/base.txt
+    #   xblock
+    #   xblock-sdk
+webob==1.8.7
+    # via
+    #   -r requirements/base.txt
+    #   xblock
+    #   xblock-sdk
+xblock[django]==4.0.1
+    # via
+    #   -c requirements/constraints.txt
+    #   -r requirements/base.txt
+    #   xblock-sdk
+xblock-sdk==0.11.0
+    # via -r requirements/test.in
+yarl==1.9.4
+    # via
+    #   -r requirements/base.txt
+    #   aiohttp
+zipp==3.19.2
+    # via
+    #   -r requirements/base.txt
+    #   importlib-metadata
+    #   importlib-resources
+
+# The following packages are considered to be unsafe in a requirements file:
+# setuptools
diff --git a/requirements/workbench.txt b/requirements/workbench.txt
new file mode 100644
index 0000000..e69de29
diff --git a/setup.py b/setup.py
new file mode 100644
index 0000000..5254944
--- /dev/null
+++ b/setup.py
@@ -0,0 +1,43 @@
+"""Setup for shortanswer XBlock."""
+
+import os
+
+from setuptools import setup
+
+
+def package_data(pkg, roots):
+    """Generic function to find package_data.
+
+    All of the files under each of the `roots` will be declared as package
+    data for package `pkg`.
+
+    """
+    data = []
+    for root in roots:
+        for dirname, _, files in os.walk(os.path.join(pkg, root)):
+            for fname in files:
+                data.append(os.path.relpath(os.path.join(dirname, fname), pkg))
+
+    return {pkg: data}
+
+
+setup(
+    name="xblock-ai-eval",
+    version="0.1",
+    description="XBlocks to write short text and code entries with AI-driven evaluation",
+    license="AGPL v3",
+    packages=[
+        "ai_eval",
+    ],
+    install_requires=[
+        "XBlock",
+        "litellm>=1.42",
+    ],
+    entry_points={
+        "xblock.v1": [
+            "shortanswer_ai_eval = ai_eval:ShortAnswerAIEvalXBlock",
+            "coding_ai_eval = ai_eval:CodingAIEvalXBlock",
+        ]
+    },
+    package_data=package_data("ai_eval", ["static", "public", "templates"]),
+)