rune/oai
1
0
Fork 0
Code Issues Pull Requests Releases 2 Activity

Initial commit

Browse Source
This commit is contained in:
Rune Olsen
2025-12-04 12:12:33 +01:00
commit 48c4be5a54
5 changed files with 526 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

BIN
.DS_Store vendored Normal file
View File

Binary file not shown.

165
README.md Normal file
View File

@@ -0,0 +1,165 @@
# OAI Chat App
A command-line interface (CLI) chat application for interacting with AI language models via OpenRouter. Seamlessly switch between 300+ models, attach files (images and text), enjoy real-time streaming responses, and manage settings persistently—all from your terminal. Perfect for developers, writers, and AI enthusiasts looking for a fast, feature-rich chat experience without leaving the command line.
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
## 🌟 Features
- **Model Flexibility**: Choose from hundreds of OpenRouter-supported models (e.g., GPT-4, Claude, Gemini) with substring-based search and selection [github.com](https://github.com/owengretzinger/gitreadme) for template-inspired ease.
- **File Attachments**: Attach images (JPEG, PNG) or plain text files directly to messages using `@path` syntax, with automatic Base64 encoding for models that support modalities.
- **Streaming Responses**: Enable real-time streaming for faster, live responses; toggle on/off via config commands (inspired by streaming in modern chat apps).
- **Persistent Configuration**: Securely store API keys, base URLs, and settings in a local SQLite database for reuse across sessions.
- **Rich UI in CLI**: Enhanced console output with Rich library panels, tables, and progress indicators for a polished terminal experience.
- **Clipboard Integration**: Copy AI responses to the clipboard with a single prompt.
- **Help Command**: Built-in `/help` with examples for all commands, making the app self-documenting and user-friendly [github.com](https://github.com/Piyush64-bit/ReadMe-Generator).
- **Error Handling**: Graceful prompts for missing configs, file issues, and model incompatibilities.
- **Cross-Platform**: Works on any OS with Python and internet access, no heavy dependencies required.
## 🖼️ Screenshots
![OAI Chat App in Action](https://via.placeholder.com/600x200?text=OAI+Chat+App+Demo)
*(Replace with actual screenshot of the CLI interface, e.g., showing model selection or a chat response.)*
## 📋 Requirements
- **Python**: 3.9 or higher
- **OpenRouter API Key**: Free tier available at [openrouter.ai](https://openrouter.ai). Models like GPT-4 may require a paid plan.
- **Dependencies**: Listed in `requirements.txt` (generate automatically with tools like `pipreqs` if needed).
- **Internet Connection**: For API calls to OpenRouter.
## 🚀 Installation
1. **Clone the Repository**:
```bash
git clone https://github.com/yourusername/oai-chat-app.git # Replace with your actual repo URL
cd oai-chat-app
```
2. **Install Dependencies**:
Use `pip` for simplicity or `uv` for faster management:
```bash
# Using pip
pip install typer rich openrouter pyperclip requests
# Using uv
uv add typer rich openrouter pyperclip requests
# Using pip and requirements.txt (recommended)
pip3 install -r requirements.txt
```
3. **Set Up Configuration**:
- The app creates a SQLite database at `~/.config/oai/oai_config.db` on first run.
- Run the app once to set your API key: `python oai_chat.py chat`.
- For API key issues, visit [openrouter.ai/settings/keys](https://openrouter.ai/settings/keys).
4. **Make Executable (Optional)**:
```bash
chmod +x oai_chat.py # For Unix-based systems
```
Your app is now ready! Run `python oai_chat.py chat` to start chatting [github.com](https://github.com/waheed444/AI_Readme_Generator) inspired by simple setup in AI tools.
## 💻 Usage
Launch the app and interact via commands or direct chat:
```bash
python oai_chat.py chat
```
### Available Commands
| Command | Description | Example |
|---------------------|-----------------------------------------------------------------------------|-------------------------------------------------------------------------|
| `/help` | Display a table with examples for all commands and features. | `/help` |
| `/model [search]` | Select or switch models; optional search for substring matching. | `/model gpt` (Filters to GPT models)<br>`/model` (Shows all models) |
| `/config api` | Update OpenRouter API key and reinitialize the client. | `/config api`<br>Enter key: sk-...<br>[Output: API key updated!] |
| `/config url` | Change the base URL for custom OpenRouter endpoints. | `/config url`<br>Enter URL: https://custom.openrouter.ai/api/v1 |
| `/config stream on/off` | Toggle response streaming (default: on for real-time chat). | `/config stream off` (For full responses at once)<br>`/config stream on`|
| `/config` | View current configurations. | `/config` (Displays table: API Key, Base URL, Streaming, Database path)|
| `exit`, `quit`, `bye` | Exit the app cleanly. | `exit` |
| Chatting with Files | Attach files using `@path` (e.g., images for vision models, text for context). | `Explain this @~/demo.png`<br>(Attaches image if model supports it) |
### Example Session
```
Welcome to OAI! Type your message, '/model [search]' to select/change model, '/config api/url/stream' to configure, '/help' for examples, or 'exit'/'quit' to end.
No model selected. Use '/model' to choose one.
You: /help
[Displays help table as above]
You: /model gpt
(Shows table of GPT models)
Enter model number: 1
Selected: GPT-4o (openai/gpt-4o)
You: Hello, how are you?
Streaming response... (Press Ctrl+C to cancel)
Hello! I'm doing great, thanks for asking. How can I help you today?
Type 'c' to copy response to clipboard, or press Enter to continue: c
Response copied to clipboard!
You: Analyze this text @~/notes.txt
Streaming response... (Press Ctrl+C to cancel)
Based on the attached file, the text discusses...
You: exit
Goodbye!
```
- **Streaming**: Enabled by default for live responses; press Ctrl+C mid-stream to cancel (supported by most providers).
- **File Limits**: Max 10MB per file; only images for vision models, plain text otherwise. Supports `@~/relative/path` or absolute paths.
- **Error Recovery**: If no models fetch initially, check your API key. Prompts guide you through setup.
For more on OpenRouter models and features (e.g., tool calling), see [openrouter.ai/docs](https://openrouter.ai/docs).
## ⚙️ Configuration
All settings are stored securely in `~/.config/oai/oai_config.db` using SQLite:
- **API Key**: Required for authentication (set via `/config api`).
- **Base URL**: Defaults to `https://openrouter.ai/api/v1` (change for custom endpoints via `/config url`).
- **Streaming**: Toggles live responses (`on` for real-time, `off` for buffered). Defaults to `on`.
- **Database Path**: Automatically created; view full path with `/config`.
Edit any setting during a session without restarting. The database persists settings for future runs.
## 🛠️ Troubleshooting
- **No Models Available?** Ensure your API key is valid and has access to models [openrouter.ai](https://openrouter.ai/docs/api-keys). Check network connectivity.
- **Streaming Issues?** Some models/providers may fall back to non-streaming. Toggle with `/config stream off`.
- **File Attachment Errors?** Confirm paths are absolute (e.g., `@/home/user/file.png`) and files are under 10MB. Unsupported MIME types will be rejected.
- **API Errors?** Re-set your key or check model compatibility (e.g., vision models for images).
- **Permission Denied?** Ensure the app has read access to attached files and the database directory.
- **Installation Fails?** Verify Python version and try `pip install --upgrade pip`. Use `uv` for dependency resolution [github.com](https://github.com/Herc-Ch/readme-generator).
- **Performance?** Large files or complex prompts may slow responses—trim inputs as needed.
If issues persist, open a GitHub issue with error logs.
## 🤝 Contributing
We welcome contributions to improve the OAI Chat App!
1. Fork the repo.
2. Create a feature branch: `git checkout -b feature/new-feature`.
3. Make changes and test thoroughly.
4. Run `pipreqs .` to update `requirements.txt` if dependencies change.
5. Submit a pull request with a clear description.
Ideas: Add conversation history, multi-turn chats, or OAuth for API keys [github.com](https://github.com/mrunankpawar/repo-readme-generator).
## 📜 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## 🙏 Acknowledgments
- Built with OpenRouter for model access [openrouter.ai](https://openrouter.ai).
- Inspired by AI README generators for documentation structure [github.com](https://github.com/Herc-Ch/readme-generator), [github.com](https://github.com/owengretzinger/gitreadme), [github.com](https://github.com/Piyush64-bit/ReadMe-Generator), [github.com](https://github.com/waheed444/AI_Readme_Generator), [github.com](https://github.com/mrunankpawar/repo-readme-generator).
- CLI framework via Typer and Rich for polished UX.
Star ⭐ if this app helps—it's open-source and free! Happy chatting!

BIN
images/screenshot_01.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 297 KiB

345
oai.py Normal file
View File

@@ -0,0 +1,345 @@
#!/usr/bin/python3 -W ignore::DeprecationWarning
import os
import requests
from pathlib import Path
from typing import Optional, List
import typer
from rich.console import Console
from rich.panel import Panel
from rich.table import Table
from rich.text import Text
from openrouter import OpenRouter
import pyperclip
import mimetypes
import base64
import re
import sqlite3 # Added for SQLite DB integration
app = typer.Typer()
console = Console()
# DB configuration - define DB file and table setup
homefilepath = Path.home()
filepath = homefilepath.joinpath('.config/oai')
database = filepath.joinpath('oai_config.db')
DB_FILE = database
def create_table_if_not_exists():
"""Ensure the config table exists and directories are created."""
os.makedirs(os.path.dirname(DB_FILE), exist_ok=True) # Create directories if needed
with sqlite3.connect(DB_FILE) as conn:
conn.execute('''CREATE TABLE IF NOT EXISTS config (
key TEXT PRIMARY KEY,
value TEXT NOT NULL
)''')
conn.commit()
def get_config(key: str) -> Optional[str]:
"""Fetch a config value from the DB."""
create_table_if_not_exists() # Ensure table exists
with sqlite3.connect(DB_FILE) as conn:
cursor = conn.execute('SELECT value FROM config WHERE key = ?', (key,))
result = cursor.fetchone()
return result[0] if result else None
def set_config(key: str, value: str):
"""Set or update a config value in the DB."""
create_table_if_not_exists() # Ensure table exists
with sqlite3.connect(DB_FILE) as conn:
conn.execute('INSERT OR REPLACE INTO config (key, value) VALUES (?, ?)', (key, value))
conn.commit()
# Load configurations from DB on startup (or set defaults if not present)
API_KEY = get_config('api_key')
OPENROUTER_BASE_URL = get_config('base_url') or "https://openrouter.ai/api/v1" # Default if not set
STREAM_ENABLED = get_config('stream_enabled') or "on" # Default to streaming on
# Fetch models once at module level (with loaded BASE_URL)
models_data = []
text_models = [] # Filtered models: allow "image" but exclude "video"
try:
headers = {"Authorization": f"Bearer {API_KEY}"} if API_KEY else {}
response = requests.get(f"{OPENROUTER_BASE_URL}/models", headers=headers)
response.raise_for_status()
models_data = response.json()["data"]
# Filter: Exclude "video" models, but allow "image" for attachments
text_models = [m for m in models_data if "modalities" not in m or "video" not in (m.get("modalities") or [])]
except Exception as e:
models_data = []
text_models = []
@app.command()
def chat():
"""Start the OAI chat app with OpenRouter models."""
global API_KEY, OPENROUTER_BASE_URL, STREAM_ENABLED # Allow updates via /config commands
if not API_KEY:
console.print("[bold red]API key not found in database. Please set it with '/config api'.[/]")
# Prompt for API key on startup if missing
try:
new_api_key = typer.prompt("Enter your API key")
if new_api_key.strip():
set_config('api_key', new_api_key.strip())
API_KEY = new_api_key.strip()
console.print("[bold green]API key saved. Re-run the app or continue with '/model'.[/]")
else:
raise typer.Exit()
except Exception:
console.print("[bold red]No API key provided. Exiting.[/]")
raise typer.Exit()
if not text_models:
console.print("[bold red]No suitable models available or error fetching models (check API key and base URL).[/]")
raise typer.Exit()
selected_model = None
client = OpenRouter(api_key=API_KEY)
console.print("[bold blue]Welcome to OAI! Type your message, '/model [search]' to select/change model, '/config api/url/stream' to configure, '/help' for examples, or 'exit'/'quit' to end.[/]")
if not selected_model:
console.print("[bold yellow]No model selected. Use '/model' to choose one.[/]")
while True:
try:
user_input = typer.prompt("You").strip()
if user_input.lower() in ["exit", "quit", "bye"]:
console.print("[bold yellow]Goodbye![/]")
return
# Handle /model command (unchanged)
if user_input.startswith("/model"):
args = user_input[7:].strip() # Get everything after "/model" as search term
search_term = args if args else ""
filtered_models = text_models
if search_term:
# Substring filter (case-insensitive) on name or id
filtered_models = [m for m in text_models if search_term.lower() in m["name"].lower() or search_term.lower() in m["id"].lower()]
if not filtered_models:
console.print(f"[bold red]No models match '{search_term}'. Try '/model' without search.[/]")
continue
# Display filtered models
table = Table("No.", "Name", "ID", show_header=True, header_style="bold magenta")
for i, model in enumerate(filtered_models, 1):
table.add_row(str(i), model["name"], model["id"])
console.print(Panel(table, title=f"[bold green]Available Models ({'All' if not search_term else f'Search: {search_term}'})[/]", title_align="left"))
# Prompt selection
while True:
try:
choice = int(typer.prompt("Enter model number (or 0 to cancel)"))
if choice == 0:
break
if 1 <= choice <= len(filtered_models):
selected_model = filtered_models[choice - 1]
console.print(f"[bold cyan]Selected: {selected_model['name']} ({selected_model['id']})[/]")
break
console.print("[bold red]Invalid choice. Try again.[/]")
except ValueError:
console.print("[bold red]Invalid input. Enter a number.[/]")
continue
# Handle /config command (updated to include stream toggle)
if user_input.startswith("/config"):
args = user_input[8:].strip().lower() # Get args after "/config"
if args == "api":
try:
new_api_key = typer.prompt("Enter new API key")
if new_api_key.strip():
set_config('api_key', new_api_key.strip())
API_KEY = new_api_key.strip()
client = OpenRouter(api_key=API_KEY) # Reinitialize client
console.print("[bold green]API key updated and client reinitialized![/]")
else:
console.print("[bold yellow]No change made.[/]")
except Exception as e:
console.print(f"[bold red]Error updating API key: {e}[/]")
elif args == "url":
try:
new_url = typer.prompt("Enter new base URL (e.g., https://openrouter.ai/api/v1)")
if new_url.strip():
set_config('base_url', new_url.strip())
OPENROUTER_BASE_URL = new_url.strip()
console.print("[bold green]Base URL updated![/]")
else:
console.print("[bold yellow]No change made.[/]")
except Exception as e:
console.print(f"[bold red]Error updating base URL: {e}[/]")
elif args.startswith("stream"):
sub_args = args[7:].strip() # After "stream"
if sub_args in ["on", "off"]:
set_config('stream_enabled', sub_args)
STREAM_ENABLED = sub_args
console.print(f"[bold green]Streaming {'enabled' if sub_args == 'on' else 'disabled'}.[/]")
else:
console.print("[bold yellow]Usage: /config stream on|off[/]")
else:
# /config with no args: Display current configs
table = Table("Setting", "Value", show_header=True, header_style="bold magenta")
table.add_row("API Key", API_KEY or "[Not set]")
table.add_row("Base URL", OPENROUTER_BASE_URL or "[Not set]")
table.add_row("Streaming", "Enabled" if STREAM_ENABLED == "on" else "Disabled")
table.add_row("Database", str(database) or "[Not set]")
console.print(Panel(table, title="[bold green]Current Configurations[/]", title_align="left"))
continue
# Handle /help command (new: display examples for each command)
if user_input.lower() == "/help":
help_table = Table("Command", "Description", "Example", show_header=True, header_style="bold cyan")
help_table.add_row(
"/model [search]",
"Select or change the model for chatting. Supports searching by name or ID.",
"/model gpt\nYou: 1\n(Selects first matching model)"
)
help_table.add_row(
"/config api",
"Set or update the OpenRouter API key.",
"/config api\nEnter new API key: sk-...\n[bold green]API key updated and client reinitialized![/bold green]"
)
help_table.add_row(
"/config url",
"Set or update the base URL for OpenRouter.",
"/config url\nEnter new base URL: https://api.example.com/v1\n[bold green]Base URL updated![/bold green]"
)
help_table.add_row(
"/config stream on|off",
"Enable or disable response streaming.",
"/config stream off\n[bold green]Streaming disabled.[/bold green]"
)
help_table.add_row(
"/config",
"View all current configurations.",
"/config\n(Displays table of API Key, Base URL, etc.)"
)
help_table.add_row(
"/help",
"Show this help menu with examples.",
"/help\n(Displays this table)"
)
help_table.add_row(
"Chatting with files",
"Attach images or plain text files to messages using '@path'.",
"Explain this @/Users/me/image.jpg\n(Attaches image.jpg if supported by the model)"
)
help_table.add_row(
"Exiting",
"Quit the chat app with either 'exit', 'quit' or 'bye'",
"exit\n[bold yellow]Goodbye![/bold yellow]"
)
console.print(Panel(help_table, title="[bold cyan]OAI Chat Help - Command Examples[/]", title_align="center"))
continue
if not selected_model:
console.print("[bold yellow]Select a model first with '/model'.[/]")
continue
# Process file attachments in the prompt (unchanged)
content_blocks = []
text_part = user_input
file_attachments = []
# Regex to find @path (e.g., @/Users/user/file.jpg or @c:\folder\file.txt)
file_pattern = r'@([^\s]+)' # @ followed by non-spaces
for match in re.finditer(file_pattern, user_input):
file_path = match.group(1)
expanded_path = os.path.expanduser(os.path.abspath(file_path))
if not os.path.exists(expanded_path) or os.path.isdir(expanded_path):
console.print(f"[bold red]File not found or is a directory: {expanded_path}[/]")
continue
file_size = os.path.getsize(expanded_path)
if file_size > 10 * 1024 * 1024: # 10MB limit
console.print(f"[bold red]File too large (>10MB): {expanded_path}[/]")
continue
mime_type, _ = mimetypes.guess_type(expanded_path)
try:
with open(expanded_path, 'rb') as f:
file_data = f.read()
if mime_type and mime_type.startswith('image/'):
if "image" not in selected_model.get("modalities", []):
console.print("[bold red]Selected model does not support image attachments.[/]")
continue
b64_data = base64.b64encode(file_data).decode('utf-8')
content_blocks.append({"type": "image_url", "image_url": {"url": f"data:{mime_type};base64,{b64_data}"}})
elif mime_type == 'text/plain':
text_content = file_data.decode('utf-8')
content_blocks.append({"type": "text", "text": text_content})
else:
console.print(f"[bold red]Unsupported file type: {mime_type} for {expanded_path}. Only images and plain text supported.[/]")
continue
file_attachments.append(file_path)
except Exception as e:
console.print(f"[bold red]Error reading file {expanded_path}: {e}[/]")
continue
# Remove @path from text_part
text_part = re.sub(file_pattern, '', text_part).strip()
# Build message content (unchanged)
if text_part or content_blocks:
message_content = []
if text_part:
message_content.append({"type": "text", "text": text_part})
message_content.extend(content_blocks)
else:
console.print("[bold red]Prompt cannot be empty.[/]")
continue
# Send to model with streaming (enable/disable based on config)
is_streaming = STREAM_ENABLED == "on"
if is_streaming:
console.print("[bold green]Streaming response... (Press Ctrl+C to cancel)[/]")
else:
console.print("[bold green]Thinking...[/]", end="\r")
try:
response = client.chat.send(
model=selected_model["id"],
messages=[{"role": "user", "content": message_content}],
stream=is_streaming # Enable/disable streaming
)
except Exception as e:
console.print(f"[bold red]Error sending request: {e}[/]")
continue
if is_streaming:
# Handle streaming response
full_response = ""
try:
for chunk in response:
# Check for mid-stream errors (per OpenRouter docs)
if 'error' in chunk and chunk.error:
console.print(f"\n[bold red]Stream error: {chunk.error.message}[/]")
break
# Print streaming content
if chunk.choices and chunk.choices[0].delta.content:
content_chunk = chunk.choices[0].delta.content
print(content_chunk, end="", flush=True)
full_response += content_chunk
if full_response:
console.print() # New line after streaming
except KeyboardInterrupt:
console.print("\n[bold yellow]Streaming cancelled![/]")
continue
else:
# Non-streaming fallback
full_response = response.choices[0].message.content if response.choices else ""
console.print(f"\r{' ' * 20}\r", end="") # Clear "Thinking..." overlay
if full_response:
# Pretty print response panel (only if we have content)
console.print(Panel(full_response, title="[bold green]AI Response[/]", title_align="left"))
# Clipboard prompt (unchanged)
try:
copy_choice = input("Type 'c' to copy response to clipboard, or press Enter to continue: ").strip().lower()
if copy_choice == "c":
pyperclip.copy(full_response)
console.print("[bold green]Response copied to clipboard![/]")
except (EOFError, KeyboardInterrupt):
pass # Handle non-interactive environments gracefully
else:
console.print("[bold red]No response received.[/]")
except Exception as e:
console.print(f"[bold red]Error: {e}[/]")
console.print("[bold yellow]Try again or use '/model' to select a different model.[/]")
if __name__ == "__main__":
app()

16
requirements.txt Normal file
View File

@@ -0,0 +1,16 @@
annotated-types==0.7.0
anyio==4.9.0
certifi==2025.1.31
distro==1.9.0
h11==0.14.0
httpcore==1.0.7
httpx==0.28.1
idna==3.10
jiter==0.9.0
openai==1.71.0
pydantic==2.11.2
pydantic_core==2.33.1
sniffio==1.3.1
tqdm==4.67.1
typing-inspection==0.4.0
typing_extensions==4.13.1