Backfire vs QuotaGuard

Backfire compared with QuotaGuard

These two get compared because both sit between your code and someone else's API. QuotaGuard answers a question about your source address: route this app through us and it will always arrive from the same IP, which you can hand to a partner who screens callers. Backfire answers a question about your destination: is this URL allowed to be called at all, and can you prove afterwards what happened.

Which one you actually need

If a partner must allowlist a source IP before they will talk to you, buy the static-IP proxy — Backfire has no IP to give and will not pretend to. If the risk is the URL rather than the caller, a fixed IP changes nothing about it, and the relay is the shorter path.

Job it does

QuotaGuard
One stable outbound IP for your app
Backfire
One guarded, metered, logged call

Exit identity

QuotaGuard
A dedicated static IP you can hand to a partner
Backfire
Shared relay infrastructure, attributable to your key

Target allowlisting

QuotaGuard
Not the product
Backfire
Every address checked after DNS, refused if private

How you call it

QuotaGuard
Your HTTP client is routed through a proxy URL
Backfire
One JSON POST to the forward endpoint

Per-call record

QuotaGuard
Request logs and quota counters
Backfire
Method, target, status, bytes, duration, error code — exportable

Refusals

QuotaGuard
Network-level failures
Backfire
A code, a reason and a log line, never charged

Contract shape

QuotaGuard
Ongoing subscription with a plan ceiling
Backfire
Prepaid trips, no subscription, no expiry

Where it fits

QuotaGuard
Runtimes that need a fixed egress address
Backfire
Any runtime that can send an HTTPS POST

Side by side

quota-guard
# Static-IP proxy: the source address is the product
HTTPS_PROXY=http://user:pass@static.proxy.example:9293   curl https://api.partner.example/v1/orders

# Backfire: the destination is the product — an IP literal is refused
POST /functions/forward
{ "url": "http://169.254.169.254/latest/meta-data/" }
# → 403 blocked_target, logged, no trip charged

Where the other one wins

  • A fixed IP answers "who is calling". It does not answer "what did we call, was it allowed, and what came back" — which is the question an incident review actually asks.
  • Backfire egresses from shared relay infrastructure. If your partner insists on a per-customer address, that is a legitimate requirement and this is the wrong product for it.
  • The two coexist comfortably: a static IP for the one partner that demands it, the relay for the URLs your users, your models and your webhooks hand you.
  • Neither replaces an egress policy inside your own VPC. That is a network control; both of these are application controls.

What Backfire will not do

  • No static, dedicated or per-customer outbound IP — and no way to pin one.
  • No HTTP CONNECT or SOCKS mode: it is a JSON forward, not a proxy URL for an HTTP client.
  • A partner that screens source addresses will refuse the relay, correctly.
  • One trip per completed roundtrip; a refusal before the call is free.

Other comparisons

Questions about this

Related reading

Try it with 5,000 free trips.

No card, no subscription, and the balance never expires.