https://github.com/furk4neg3/jotform-test-ai-agent-changes
Automated live testing for Jotform AI Agent updates—no manual previews required.
https://github.com/furk4neg3/jotform-test-ai-agent-changes
ai-agent flask open-ai python
Last synced: 3 months ago
JSON representation
Automated live testing for Jotform AI Agent updates—no manual previews required.
- Host: GitHub
- URL: https://github.com/furk4neg3/jotform-test-ai-agent-changes
- Owner: furk4neg3
- Created: 2025-07-08T12:31:57.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-07-09T14:11:18.000Z (3 months ago)
- Last Synced: 2025-07-09T15:20:48.602Z (3 months ago)
- Topics: ai-agent, flask, open-ai, python
- Language: Python
- Homepage:
- Size: 21.5 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Jotform AI Agent Auto-Preview
Automatically test your Jotform AI Agent immediately after any configuration change—no more manual “Preview” clicks.
---
## 🚀 Features
* **Add Knowledge**
Push new knowledge materials (text) to your agent and instantly generate a test question to verify integration.
* **Add Action**
Define triggers (e.g. “User Talks About”, “Sentence Contains”) and corresponding actions (e.g. “Say Exact Message”) and automatically validate them via a generated prompt.
* **Change Persona**
Update agent properties—name, role, language, tone, chattiness, or add chat guidelines—and see live conversational previews.
* **Unified UI**
A single-page interface to configure your agent, powered by Flask and a React-like vanilla JS front-end.
---
## 📦 Tech Stack
* **Backend**: Flask (Python)
* **Custom HTTP Client:** Reverse-engineered Jotform’s hidden AI Agent web API via browser DevTools (network tab & cURL) to implement all agent-builder calls from scratch
* **Jotform Client**: `JotformAIAgentClient` wrapping Jotform AI Agent Builder & Chat APIs
* **AI Prompting**: OpenAI GPT (via `openai` Python SDK)
* **Frontend**: Static HTML/CSS/JS (`index.html`) served from `static/`
* **Environment Management**: python-dotenv
---
## 🛠️ Getting Started
### Prerequisites
* Python 3.8+
* A Jotform **Session Token** (for authenticated API calls)
* An **OpenAI API Key**
### Installation
1. **Clone the repo**
```bash
git clone https://github.com/your-org/jotform-ai-preview.git
cd jotform-ai-preview
```
2. **Install dependencies**
```bash
pip install -r requirements.txt
```
3. **Create `.env`**
```ini
# .env
JOTFORM_SESSION=your_jotform_session_cookie
OPENAI_API_KEY=your_openai_api_key
```
### Running the App
```bash
python app.py
```
By default, Flask will serve on `http://localhost:5000/`. Open that in your browser to load the configuration UI.
---
## 🗂️ Project Structure
```
.
├── app.py # Flask app defining all endpoints (/add_knowledge, /add_action, etc.)
├── jotform_client.py # JotformAIAgentClient: wrappers for materials, tasks, properties, chats
├── requirements.txt # Python dependencies
├── .env # Environment variables (not committed)
└── static/
└── index.html # Single-page UI for agent configuration and live preview
```
---
## 🔌 API Endpoints
| Endpoint | Method | Description |
| ----------------- | ------ | ---------------------------------------------------------------- |
| `/` | GET | Serve the static configuration UI |
| `/add_knowledge` | POST | Add new knowledge and preview via generated test prompt |
| `/add_action` | POST | Add new action and preview via generated test prompt |
| `/update_name` | POST | Rename agent, then preview |
| `/update_role` | POST | Change agent role, then preview |
| `/add_guideline` | POST | Append a chat guideline to persona, then preview |
| `/update_persona` | POST | Update any persona property (language, tone, etc.), then preview |
---
## 💡 How It Works
1. **Configuration** (UI): you enter your **Agent ID** and configure knowledge, actions, or persona.
2. **Backend Flow** (e.g. in `/add_knowledge`):
* Calls `client.add_knowledge(...)` to push new material.
* Uses OpenAI to generate a single test prompt (`generate_test_prompt`).
* Invokes `client._preview_change(...)` to start a fresh chat and run the prompt, capturing both greeting and agent reply.
3. **Results Display**: The UI parses the JSON and renders:
* **Initial Greeting**
* **Test Prompt**
* **Agent Response**
* **Raw API Result**
---
## 🤝 Contributing
1. Fork and create a branch (`git checkout -b feature/foo`)
2. Commit your changes (`git commit -am 'Add some foo'`)
3. Push to the branch (`git push origin feature/foo`)
4. Open a Pull Request
---
## 📄 License
This project is licensed under the MIT License.
---
*Happy testing—no more manual previews!*