axiomax-esg-sdk 1.0.0

Official client SDK for AXIOMAX ESG Carbon Shield. Cryptographically attest your AI inference workload carbon footprint with one line of code. Patent Pending USPTO 64/081,419.

axiomax-esg-sdk

Official client SDK for AXIOMAX ESG Carbon Shield. Cryptographically attest your AI inference workload carbon footprint with one line of code.

Patent Pending USPTO Application 64/081,419 (filed June 3, 2026). Operated by AXIOMAX LLC, Salinas, Puerto Rico.

Install

pip install axiomax-esg-sdk

Quick start with OpenAI

from axiomax_esg_sdk import AxiomaxClient, wrap_openai
import openai

axe = AxiomaxClient(license_key="axe_lic_2026_acme_...")

# One-line wrap — every chat completion now cryptographically attested
client = wrap_openai(openai.OpenAI(api_key="..."), axe)

# Use OpenAI normally
response = client.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": "hello"}],
)

# Periodically issue a signed token from accumulated records
token = axe.issue_token()
print(token["signature_ed25519"])

Quick start with Anthropic Claude

from axiomax_esg_sdk import AxiomaxClient, wrap_anthropic
import anthropic

axe = AxiomaxClient(license_key="axe_lic_...")
client = wrap_anthropic(anthropic.Anthropic(api_key="..."), axe)

response = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=1024,
    messages=[{"role": "user", "content": "hello"}],
)

Quick start with AWS Bedrock

import boto3
from axiomax_esg_sdk import AxiomaxClient, wrap_bedrock

axe = AxiomaxClient(license_key="axe_lic_...")
bedrock = wrap_bedrock(boto3.client("bedrock-runtime"), axe)

# every invoke_model is now attested (duration only — parse body for token counts)
response = bedrock.invoke_model(modelId="anthropic.claude-3-sonnet-20240229-v1:0", body=...)

Manual recording (any provider)

from axiomax_esg_sdk import AxiomaxClient

axe = AxiomaxClient(license_key="axe_lic_...", soc_profile="apple_m4_pro", region="us_east")

# After your inference completes, record metrics:
axe.record_inference(
    model_id="gpt-4",
    tokens_in=120,
    tokens_out=340,
    duration_ms=1800,
)

# Sign accumulated batch
token = axe.issue_token()

Auto-batching every N seconds

axe = AxiomaxClient(
    license_key="axe_lic_...",
    batch_interval_sec=3600,  # auto-issue token every hour
)

Verification

Every issued token is verifiable by any third party at https://verify.axiomaxllc.com or using our open-source verifier pip install axiomax-esg-verify.

License

MIT for this SDK. The AXIOMAX ESG Carbon Shield system architecture, calibration coefficients, and brand are proprietary to AXIOMAX LLC (Patent Pending USPTO 64/081,419).

Links

• Main: https://axiomaxllc.com
• Public verifier: https://verify.axiomaxllc.com
• Pricing: https://axiomaxllc.com/pricing.html
• Open-source verifier spec: https://github.com/axiomaxllc/esg-carbon-shield