amosgyamfi

import asyncio
import logging
from uuid import uuid4

from dotenv import load_dotenv

from vision_agents.core.edge.types import User
from vision_agents.core.agents import Agent
from vision_agents.plugins import getstream, deepgram, smart_turn, gemini, cartesia

amosgyamfi

import asyncio
import logging
from uuid import uuid4

from dotenv import load_dotenv

from vision_agents.core.edge.types import User
from vision_agents.core import agents
from vision_agents.plugins import getstream, ultralytics, gemini, openai

amosgyamfi

Yoga AI Voice Instructor Guide

1. Overview

You are a voice yoga instructor — an expert in the art and science of yoga postures (asanas) as defined in the Pocket Yoga Pose Library.

You use YOLO pose analysis to see the practitioner’s exact movements — body alignment, angles, and transitions. Your job is to observe, assess, and guide users through precise, safe, and mindful practice.

Voice and Personality • Speak only in English and with a female voice and a soft American accent — grounding, witty, and slightly snarky when correcting poor form.

amosgyamfi

"""GitHub MCP Demo - Demonstrate function calling with OpenAI Realtime and GitHub MCP.

This demo shows how OpenAI Realtime can use GitHub MCP tools for real-time function calling
during live conversations. The agent can interact with GitHub repositories, issues, and more
using voice commands through the OpenAI Realtime API.
"""

import asyncio
import logging
import os

amosgyamfi

import asyncio
import logging
from uuid import uuid4

from dotenv import load_dotenv

from vision_agents.core.edge.types import User
from vision_agents.core.agents import Agent
from vision_agents.plugins import fish, getstream, deepgram, smart_turn, xai

amosgyamfi

import asyncio
import logging
from uuid import uuid4

from dotenv import load_dotenv

from vision_agents.core.edge.types import User
from vision_agents.core.agents import Agent
from vision_agents.plugins import cartesia, getstream, deepgram, smart_turn, gemini

amosgyamfi

import asyncio
from uuid import uuid4

from vision_agents.core.edge.types import User
from vision_agents.core import agents
from vision_agents.plugins import openai, getstream

async def start_agent() -> None:
    # create an agent to run with Stream's edge, openAI realtime llm
    agent = agents.Agent(

amosgyamfi

import asyncio
import logging
from uuid import uuid4

from dotenv import load_dotenv

from vision_agents.core.edge.types import User
from vision_agents.plugins import getstream, gemini 
from vision_agents.core import agents, cli

amosgyamfi

import io
import json
import re
from typing import Any, Dict, List

import streamlit as st
from google import genai
from google.genai import types
from PIL import Image, ImageDraw, ImageFont

amosgyamfi

from google import genai
from google.genai import types

from PIL import Image

# Initialize the GenAI client and specify the model
MODEL_ID = "gemini-robotics-er-1.5-preview"
PROMPT = """
          Point to no more than 10 items in the image. The label returned
          should be an identifying name for the object detected.