Developer transport layer

Residential proxy API access through standard HTTP and SOCKS5 clients

Build proxy routing into jobs with explicit credentials, geography, session policy, timeouts, and response classification—without coupling application logic to a custom scraping SDK.

proxy-session.log

$ configure route

country=US mode=rotating timeout=15s

$ validate response

route

matched

status

200

bytes

tracked

✓ bounded retry policy active

Start with a small authenticated request

Keep credentials in environment variables and validate the observed response before adding retries or concurrency.

proxy-check.shbash
curl --proxy "http://{BF_PROXY_USER}:{BF_PROXY_PASS}@p1.bytesflows.com:8001" \
  --connect-timeout 10 \
  --max-time 30 \
  --fail-with-body \
  https://httpbin.org/ip

A successful proxy connection does not prove the target page is usable. Validate target status and expected content separately.

An API integration lifecycle that stays observable

Treat credentials, route selection, target response, and usage accounting as separate concerns.

STEP 01

Create a route profile

Define endpoint, protocol, geography, rotating or sticky behavior, timeout, and allowed destination policy.

STEP 02

Send an authenticated request

Pass credentials through the client’s proxy settings and keep them out of URLs stored in logs or error reports.

STEP 03

Classify the result

Separate proxy authentication, connection, timeout, target status, invalid content, and policy denial.

STEP 04

Record usage and retry

Track bytes and retry reason, then retry only temporary classes within a bounded budget.

Residential proxy API FAQ

Is this a scraping API that returns parsed JSON?+

No. It is standard proxy transport. Your application remains responsible for the HTTP request, browser execution, validation, parsing, storage, and compliance rules.

What is the difference between rotation and sticky sessions?+

Rotation is suitable for independent requests. Sticky sessions keep one route for a limited multi-step workflow and should use a non-sensitive session identifier.

How should credentials be stored?+

Use a secret manager or protected environment variables. Redact usernames and passwords from traces, screenshots, analytics, and support output.

Which failures should be retried?+

Retry a small number of temporary connection, timeout, or server failures. Do not blindly retry invalid credentials, explicit denial, or repeated challenge pages.

Controlled evaluation

Validate the proxy interface before building a worker pool

Confirm authentication, route selection, target validation, retry classes, and byte accounting with a small request set.