Apirelio Flask SDK

Documentation · PyPI · Apirelio

Connect Flask errors, latency and releases to affected customers without capturing request or response payloads.

Native Flask request-lifecycle instrumentation for privacy-safe, customer-aware API analytics. It records the resolved URL rule and only enqueues after Flask has produced the response.

pip install apirelio-flask

import os

from flask import Flask, g
from apirelio_flask import install_apirelio

app = Flask(__name__)

install_apirelio(
    app,
    api_key=os.environ.get("APIRELIO_API_KEY", ""),
    endpoint=os.environ.get("APIRELIO_ENDPOINT", "https://apirelio.com"),
    service="billing-api",
    environment="production",
    include_routes=("/api/**",),
    exclude_routes=("/api/health", "/api/internal/**"),
    resolve_customer=lambda request: {
        "id": str(g.account.id),
        "name": g.account.name,
        "plan": g.account.plan,
    } if getattr(g, "account", None) else None,
)

Authentication must run before Apirelio's after_request resolver executes. Use get_request_context() inside a view to attach allow-listed scalar metadata or a stable error code.

The extension stores itself at app.extensions["apirelio"]. Bodies, query strings, cookies, authorization values, email addresses and IP addresses are never captured. Python 3.9+, Flask 2.3 and Flask 3.x are supported.