anosys-logger-4-openai 0.0.11

A helper package, to support implementation to AnoSys platform easly

AnoSys Logger for OpenAI - Python

Automatically capture and send OpenAI API calls to AnoSys for monitoring, analytics, and observability.

Features

✨ Automatic OpenAI Instrumentation - Captures all OpenAI API calls via OpenTelemetry
✨ Streaming Support - Detects and logs streaming responses
✨ Custom Function Decorators - Log any Python function (sync or async)
✨ OpenTelemetry Semantic Conventions - Follows Gen AI standards
✨ Error Tracking - Captures exceptions with full stack traces
✨ Zero Configuration - Works out of the box with just your API key

Installation

pip install anosys-logger-4-openai

Quick Start

1. Get Your AnoSys API Key

Visit https://console.anosys.ai/collect/integrationoptions to get your API key.

2. Basic Usage with OpenAI

import os
from openai import OpenAI
from AnosysLoggers import AnosysOpenAILogger

# Set your API keys
os.environ["OPENAI_API_KEY"] = "your-openai-api-key"
os.environ["ANOSYS_API_KEY"] = "your-anosys-api-key"

# Initialize AnoSys logger (do this once)
AnosysOpenAILogger()

# Use OpenAI as normal - all calls are automatically logged
client = OpenAI()

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain why AnoSys is great for AI observability"}
    ]
)

print(response.choices[0].message.content)

That's it! All your OpenAI calls are now being sent to AnoSys. 🎉

Advanced Usage

Streaming Responses

Streaming is automatically detected and logged:

from openai import OpenAI
from AnosysLoggers import AnosysOpenAILogger

AnosysOpenAILogger()
client = OpenAI()

stream = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Write a haiku about AI"}],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

The complete aggregated response will be logged to AnoSys.

Custom Function Decorators

Log any Python function (sync or async):

from AnosysLoggers import anosys_logger

@anosys_logger(source="my_app.calculations")
def calculate_score(data):
    # Your function logic
    return sum(data) / len(data)

# Function calls are automatically logged
result = calculate_score([85, 90, 78, 92])

Async Functions:

@anosys_logger(source="my_app.async_tasks")
async def fetch_data(url):
    # Your async logic
    return await some_async_operation()

# Async calls are also logged
result = await fetch_data("https://api.example.com")

Raw Logger

Send custom data directly:

from AnosysLoggers import anosys_raw_logger

# Log any custom data
anosys_raw_logger({
    "event": "user_action",
    "action": "button_click",
    "timestamp": "2024-01-15T10:30:00Z",
    "user_id": "12345"
})

Custom Configuration

from AnosysLoggers import setup_api

# Use a custom endpoint (advanced)
setup_api(path="https://custom.anosys.endpoint")

# Or with custom index starting points (rarely needed)
setup_api(starting_indices={
    "string": 200,
    "number": 10,
    "bool": 5
})

What Data is Captured?

OpenTelemetry Semantic Conventions

Following the OpenTelemetry Gen AI standards:

• gen_ai.system - Always "openai"
• gen_ai.request.model - Model requested (e.g., "gpt-4o-mini")
• gen_ai.response.model - Model that responded
• gen_ai.request.temperature - Temperature parameter
• gen_ai.request.max_tokens - Max tokens parameter
• gen_ai.request.top_p - Top-p parameter
• gen_ai.response.finish_reasons - Why the response ended
• gen_ai.usage.input_tokens - Input token count
• gen_ai.usage.output_tokens - Output token count

Additional Fields

• Request/response messages
• Timestamps and duration
• Error details (if any)
• Trace IDs for distributed tracing
• Custom metadata

Error Handling

Errors are automatically captured with full context:

@anosys_logger(source="my_app.risky_function")
def risky_operation():
    raise ValueError("Something went wrong")

try:
    risky_operation()
except ValueError:
    pass  # Error is still logged to AnoSys with stack trace

Environment Variables

Variable	Required	Description
ANOSYS_API_KEY	Yes	Your AnoSys API key
OPENAI_API_KEY	Yes	Your OpenAI API key
ANOSYS_DEBUG_LOGS	No	Set to false to disable raw span logging

Requirements

• Python 3.9 - 3.12
• OpenAI Python SDK
• OpenTelemetry SDK
• traceai-openai

Troubleshooting

No data appearing in AnoSys?

1. Check your API key: Ensure ANOSYS_API_KEY is set correctly
2. Initialize before OpenAI calls: Call AnosysOpenAILogger() before making OpenAI requests
3. Check network: Ensure you can reach https://api.anosys.ai

Import errors?

Make sure all dependencies are installed:

pip install --upgrade anosys-logger-4-openai

Support

• 📧 Email: [e-mail]
• 🌐 Website: https://anosys.ai
• 📚 Console: https://console.anosys.ai

License

MIT License - see LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.