jeni-py 0.1.0

Jeni

Overview • Features • Modes • Requirements • Setup • Installation • Run • Configuration • Architecture • Capabilities •

Jeni is a Python and Streamlit application for building, running, and managing Gemini-powered analytical assistants. It supports text generation, image generation and analysis, image editing, audio transcription, audio translation, text-to-speech, embeddings, document question answering, Gemini file operations, file-search stores, Google Cloud bucket management, prompt engineering, SQLite-backed data management, and export workflows.

Jeni is designed for federal data analysis, budget execution support, document review, knowledge retrieval, prompt management, and multimodal artificial intelligence experimentation.

🎥 Demo

☁️ Cloud

🧰 Overview

• Core classes imported by the application include:

✨ Features

• Gemini-first interface for text, image, audio, document, embedding, file, and cloud workflows.
• Single Streamlit application with explicit mode selection in the sidebar.
• Text generation controls for model selection, temperature, Top-P, Top-K, frequency penalty, presence penalty, output format, response schema, safety profile, stop sequences, stream mode, Google Search grounding, and URL context.
• Image workflows for generation, analysis, and editing with aspect ratio, output MIME type, response modality, Google Search grounding, and image-search options where supported.
• Audio workflows for transcription, translation, and text-to-speech using uploaded audio files or recorded browser audio.
• Document Q&A with local document loading, text extraction, chunking, embeddings, and SQLite vector retrieval with fallback cosine similarity.
• Embedding mode with text normalization, chunking, token metrics, vector display, and configurable embedding dimensions and encoding format.
• Files API mode for server-side file upload and file operations.
• File Search Stores mode for creating, retrieving, deleting, and uploading files to file-search stores.
• Google Cloud Buckets mode for creating, retrieving, deleting, and uploading files to cloud bucket-backed workflows.
• Prompt Engineering mode backed by a local SQLite Prompts table.
• Data Export mode for exporting prompt and data assets.
• Data Management mode for SQLite import, browsing, CRUD operations, profiling, filtering, aggregation, visualization, administration, and safe SQL queries.
• Footer status bar showing provider, mode, model, and active runtime settings.

🧩 Application Modes

🛠️ Requirements

🔑 API Key Setup

Jeni reads API and cloud configuration from config.py, environment variables, and Streamlit session state. Sidebar-entered values override configuration defaults for the current session.

Helpful setup references:

• Gemini API Key

📦 Installation

1. Clone the Repository

git clone https://github.com/is-leeroy-jenkins/Jeni.git
cd Jeni

2. Create and Activate a Virtual Environment

Windows PowerShell:

python -m venv .venv
.venv\Scripts\Activate.ps1

Command Prompt:

python -m venv .venv
.venv\Scripts\activate.bat

macOS / Linux:

python -m venv .venv
source .venv/bin/activate

3. Install Dependencies

python -m pip install --upgrade pip
python -m pip install -r requirements.txt

⚙️ Configuration

Set the required values in config.py, environment variables, or the Streamlit sidebar.

Example environment variables:

export GEMINI_API_KEY="your-gemini-api-key"
export GOOGLE_API_KEY="your-google-api-key"
export GOOGLE_CSE_ID="your-google-custom-search-id"
export GOOGLEMAPS_API_KEY="your-google-maps-api-key"
export GOOGLE_CLOUD_PROJECT_ID="your-google-cloud-project-id"
export GOOGLE_CLOUD_LOCATION="us-central1"

Windows PowerShell:

setx GEMINI_API_KEY "your-gemini-api-key"
setx GOOGLE_API_KEY "your-google-api-key"
setx GOOGLE_CSE_ID "your-google-custom-search-id"
setx GOOGLEMAPS_API_KEY "your-google-maps-api-key"
setx GOOGLE_CLOUD_PROJECT_ID "your-google-cloud-project-id"
setx GOOGLE_CLOUD_LOCATION "us-central1"

🚀 Running the Streamlit Application

From the project root:

streamlit run app.py

Once running, the application is available at:

http://localhost:8501

🧠 Text Generation

The Text mode provides Gemini chat and text generation through the Chat wrapper.

Supported controls include:

📷 Images

The Images mode supports three tabs:

Image controls include:

• Image mode: Generation, Analysis, or Editing
• Model selection by workflow type
• Temperature and Top-P
• Max output tokens
• Candidate count
• Response mode
• Output MIME type
• Aspect ratio
• Image size where supported
• Google Search grounding where supported
• Google Image Search where supported

🎧 Audio

The Audio mode supports:

Audio controls include:

• Task selection
• Model selection
• Language selection
• Voice selection for TTS
• Sample rate
• Output format
• Temperature
• Top-P
• frequency penalty
• presence penalty
• loop and autoplay
• start and end time
• max output tokens
• system prompt template support

🔢 Embeddings

The Embedding mode supports:

• Embedding model selection
• Encoding format selection
• Dimension selection
• Chunk-size control
• Chunk-overlap control
• Text normalization
• Text chunking
• Embedding generation
• Token, word, unique-word, type-token ratio, and character metrics
• Vector display in a Streamlit data editor

📓 Document Q&A

The Document Q&A mode supports local document upload and retrieval-augmented question answering.

Supported document loading includes:

• pdf
• txt
• md
• docx

The workflow includes:

1. Upload a document.
2. Preview the document when possible.
3. Extract text from the file.
4. Normalize and chunk the text.
5. Generate embeddings using sentence-transformers.
6. Store vectors in sqlite-vec when available.
7. Fall back to in-memory cosine similarity when vector-table support is unavailable.
8. Retrieve relevant chunks.
9. Build a document-grounded prompt.
10. Return an answer through the chat pipeline.

📚 Files API

The Files mode exposes Gemini file-oriented workflows through the Files wrapper.

Common workflows include:

• File upload
• File ID tracking
• File metadata review
• File purpose/type management
• File-backed prompt workflows where supported by the selected model

📦 File Search Stores

The File Search Stores mode supports file-search store management through the FileSearch wrapper.

Supported workflows include:

Supported upload types include:

• pdf
• txt
• md
• docx
• png
• jpg
• jpeg

🧊 Google Cloud Buckets

The Google Cloud Buckets mode supports cloud bucket management through the CloudBuckets wrapper.

Supported workflows include:

📝 Prompt Engineering

The Prompt Engineering mode manages reusable prompts stored in the local SQLite Prompts table.

Prompt records include:

Prompt Engineering supports:

• Prompt search
• Prompt sorting
• Prompt pagination
• Prompt selection
• Prompt editing
• Prompt insertion
• Prompt update
• Prompt deletion
• Cascading selected prompts into system instructions

📭 Data Export

The Data Export mode supports application export workflows, including prompt/system-instruction export and local data export where configured.

🏛️ Data Management

The Data Management mode provides a SQLite administration and exploration interface.

Tabs include:

Visualization options include:

• Histogram
• Bar chart
• Line chart
• Scatter plot
• Box plot
• Pie chart
• Correlation heatmap

SQL execution is guarded by a read-only validator that allows SELECT, WITH, EXPLAIN, and read-oriented PRAGMA statements while blocking destructive operations such as INSERT, UPDATE, DELETE, DROP, ALTER, CREATE, ATTACH, DETACH, VACUUM, REPLACE, and TRIGGER.

🧩 Design and Architecture

Jeni uses a traditional layered Streamlit architecture:

💻 Capabilities

📁 File Organization

🧪 Example Usage

Text Generation

from gemini import Chat

chat = Chat()
response = chat.generate_text(
    prompt="Explain how random forests reduce overfitting.",
    model="gemini-2.0-flash"
)

print(response)

Embeddings

from gemini import Embeddings

embedding = Embeddings()
vectors = embedding.create(
    text=["Federal budget execution requires accurate obligations tracking."],
    model="text-embedding-004",
    task_type="RETRIEVAL_DOCUMENT"
)

print(vectors)

Image Generation

from gemini import Images

images = Images()
result = images.generate(
    prompt="A clean technical diagram of a retrieval augmented generation pipeline.",
    model="gemini-2.5-flash-image-preview"
)

print(result)

Audio Transcription

from gemini import Transcription

transcriber = Transcription()
text = transcriber.transcribe(
    "audio/meeting.m4a",
    model="gemini-2.0-flash"
)

print(text)

Text-to-Speech

from gemini import TTS

tts = TTS()
audio_bytes = tts.create_speech(
    "Hello from Jeni.",
    model="gemini-2.5-flash-preview-tts",
    voice="Kore"
)

print(type(audio_bytes))

🚀 Application Badges

📝 Notes

• Some modes depend on model availability and the options exposed by gemini.py and config.py.
• Some Google Cloud features require valid project and location settings.
• PDF extraction depends on PyMuPDF availability.
• Vector storage uses sqlite-vec when available and falls back to cosine similarity when needed.
• The Streamlit sidebar can override configured API keys for the current session.
• The application stores local prompts and data in SQLite.

📝 License

Jeni is published under the MIT License.

PyPI application

python -m pip install jeni-py
jeni

jeni opens the tabless Windows WebView2 window by default. jeni-browser opens standard browser mode. Python 3.11–3.12 is supported. The distribution is jeni-py; the import is jeni. Set GEMINI_API_KEY before using provider features. Windows requires WebView2 Runtime. See installation and publication.