Skip to content
DeepIntShield DeepIntShield

Overview

Overview

Section titled “Overview”

DeepIntShield exposes an Anthropic-compatible route. Point a supported Anthropic SDK at it to retain native Messages API objects. Provider SDK exceptions own HTTP-error decoding, so inspect status and the structured gateway body rather than assuming upstream error text is unchanged.

You keep your Anthropic SDK-based architecture while gaining DeepIntShield features like governance, load balancing, semantic caching, and multi-provider support.

Endpoint: /anthropic

The Python shield.anthropic() builder also installs prompt-cache markers when it owns the HTTP client. See Python SDK providers and error codes.

Setup

Section titled “Setup”

Install with the Anthropic extra:

Terminal window
pip install "deepintshield[anthropic]"
from deepintshield import DeepintShield


shield = DeepintShield(virtual_key="sk-ds-your-virtual-key")
client = shield.anthropic()  # pre-wired anthropic.Anthropic


response = client.messages.create(
    model="anthropic/claude-sonnet-4-5",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello!"}],
)


print(response.content[0].text)

Provider/Model Usage Examples

Section titled “Provider/Model Usage Examples”

Use multiple providers through the same Anthropic SDK format by prefixing model names with the provider:

import anthropic


client = anthropic.Anthropic(
    base_url="https://app.deepintshield.com/anthropic",
    api_key="sk-ds-your-virtual-key",
    default_headers={"x-deepintshield-vk": "sk-ds-your-virtual-key"},
)


# Anthropic models (default)
anthropic_response = client.messages.create(
    model="claude-3-5-sonnet",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello from Claude!"}]
)


# OpenAI models via Anthropic SDK format
openai_response = client.messages.create(
    model="openai/gpt-4o-mini",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello from OpenAI!"}]
)


# Google Vertex models via Anthropic SDK format
vertex_response = client.messages.create(
    model="vertex/gemini-2.5-flash",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello from Gemini!"}]
)


# Azure models
azure_response = client.messages.create(
    model="azure/gpt-4o",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello from Azure!"}]
)


# Local Ollama models
ollama_response = client.messages.create(
    model="ollama/llama3.1:8b",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello from Ollama!"}]
)

Adding Custom Headers

Section titled “Adding Custom Headers”

Pass custom headers required by DeepIntShield plugins (like governance, telemetry, etc.):

import anthropic


client = anthropic.Anthropic(
    base_url="https://app.deepintshield.com/anthropic",
    api_key="dummy-key",
    default_headers={
        "x-deepintshield-vk": "sk-ds-your-virtual-key",  # Virtual key for governance
    }
)


response = client.messages.create(
    model="claude-3-5-sonnet",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello with custom headers!"}]
)

Using Direct Keys

Section titled “Using Direct Keys”

Pass API keys directly in requests to bypass DeepIntShield’s load balancing. You can pass any provider’s API key (OpenAI, Anthropic, Mistral, etc.) since DeepIntShield only looks for Authorization or x-api-key headers. This requires the Allow Direct API keys option to be enabled in DeepIntShield configuration.

Learn more: See Key Management for enabling direct API key usage.

import anthropic


# Using Anthropic's API key directly
client_with_direct_key = anthropic.Anthropic(
    base_url="https://app.deepintshield.com/anthropic",
    api_key="sk-your-anthropic-key"  # Anthropic's API key works
)


anthropic_response = client_with_direct_key.messages.create(
    model="claude-3-5-sonnet",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello from Claude!"}]
)


# or pass different provider keys per request using headers
client = anthropic.Anthropic(
    base_url="https://app.deepintshield.com/anthropic",
    api_key="dummy-key"
)


# Use Anthropic key for Claude
anthropic_response = client.messages.create(
    model="claude-3-5-sonnet",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello Claude!"}],
    extra_headers={
        "x-api-key": "sk-ant-your-anthropic-key"
    }
)


# Use OpenAI key for GPT models
openai_response = client.messages.create(
    model="openai/gpt-4o-mini",
    max_tokens=1000,
    messages=[{"role": "user", "content": "Hello GPT!"}],
    extra_headers={
        "Authorization": "Bearer sk-your-openai-key"
    }
)

Async Inference

Section titled “Async Inference”

Submit inference requests asynchronously and poll for results later using the x-deepintshield-async header. This is useful for long-running requests where you don’t want to hold a connection open. See Async Inference for full details.

Messages

Section titled “Messages”
import anthropic
import time


client = anthropic.Anthropic(
    base_url="https://app.deepintshield.com/anthropic",
    api_key="sk-ds-your-virtual-key",
    default_headers={"x-deepintshield-vk": "sk-ds-your-virtual-key"},
)


# Submit async request
initial = client.messages.create(
    model="anthropic/claude-sonnet-4-20250514",
    max_tokens=256,
    messages=[{"role": "user", "content": "Tell me a short story."}],
    extra_headers={"x-deepintshield-async": "true"}
)


# If content is present, the request completed synchronously
if initial.content:
    print(initial.content[0].text)
else:
    # Poll until completed
    while True:
        time.sleep(2)
        poll = client.messages.create(
            model="anthropic/claude-sonnet-4-20250514",
            max_tokens=256,
            messages=[{"role": "user", "content": "Tell me a short story."}],
            extra_headers={"x-deepintshield-async-id": initial.id}
        )
        if poll.content:
            print(poll.content[0].text)
            break

Async Headers

Section titled “Async Headers”
HeaderDescription
x-deepintshield-async: trueSubmit the request as an async job. Returns immediately with a job ID.
x-deepintshield-async-id: <job-id>Poll for results of a previously submitted async job.
x-deepintshield-async-job-result-ttl: <seconds>Override the default result TTL (default: 3600s).

Supported Features

Section titled “Supported Features”

The Anthropic integration supports all features that are available in both the Anthropic SDK and DeepIntShield core functionality. If the Anthropic SDK supports a feature and DeepIntShield supports it, the integration will work seamlessly.

Next Steps

Section titled “Next Steps”