Getting Started

Drive the Firewall
Like Any Other API.

Neurowall exposes a REST API (and a gRPC alternative) for rules, threat feeds, DDoS config, and network settings — so firewall management fits into the same CI/CD, IaC, and scripting workflows as the rest of your infrastructure.

Why It Matters

Firewall changes,
without the ticket queue.

Traditional firewall appliances gate every change behind a GUI and a change-control ticket. Neurowall's API-first design means rule changes, blocklist updates, and DDoS threshold tuning can be driven the same way you already manage the rest of your infrastructure — version-controlled, reviewed in a pull request, and applied by a pipeline.

RBAC-Gated Endpoints

Three roles — admin, operator, viewer — control who can read vs. mutate state, so automation accounts get exactly the access they need.

API Keys or JWT

Scoped API keys (X-API-Key header) skip the login round-trip entirely for automation — or use standard JWT bearer tokens if a session-based flow fits your setup better. Both work with any HTTP client, CI runner, or IaC provider without custom SDKs.

Simulate Before You Commit

The policy trace/simulation endpoints test a rule change's before/after impact — including an optional real-kernel verification path — before it touches production traffic.

REST or gRPC

Use REST for scripting and CI simplicity, or the gRPC service with TLS for high-performance integrations.

Quick Start

From zero to a
scripted rule change.

01
Authenticate

Issue a scoped API key from the admin UI or database-backed key store, and store it as a CI secret — never in a script or repo. No login round-trip needed for automation.

# set once as a CI secret, reused by every call below API_KEY="$NEUROWALL_API_KEY"
02
Read current rules

List active rules before making a change — useful as a pre-change diff in CI.

curl -s https://firewall.example.com/api/v1/rules/active \ -H "X-API-Key: $API_KEY" | jq .
03
Simulate the change

Run the proposed rule through policy trace simulation before it's applied — catch conflicts and shadowed rules early.

curl -s -X POST https://firewall.example.com/api/v1/trace/policy-simulation \ -H "X-API-Key: $API_KEY" -H "Content-Type: application/json" \ -d @proposed-rule.json | jq .
04
Apply the rule

Create the rule via the CRUD endpoint. The change syncs to the fast filter and firewall rules automatically.

curl -s -X POST https://firewall.example.com/api/v1/rules \ -H "X-API-Key: $API_KEY" -H "Content-Type: application/json" \ -d @new-rule.json | jq .
05
Verify enforcement

Fetch the single-rule view to confirm which enforcement layers actually verified the change.

curl -s https://firewall.example.com/api/v1/rules/{id} \ -H "X-API-Key: $API_KEY" | jq '.verified_layers'
Endpoint Groups

Everything the UI does,
the API does too.

GroupCovers
RulesCRUD on firewall rules, active-rule listing, kernel handle lookups
Trace & SimulationBefore/after policy simulation across both enforcement layers, optional live verification
Threat IntelBlocked-IP management, threat feed status
DDoS ConfigRate limit thresholds, protection module toggles
NetworkInterfaces, routes, WAN failover/load-balance policy, NAT intents
AuditImmutable audit log queries and retention control
Rule ExportNIST SP 800-41 rule-review column format for compliance reporting
What Other Firewalls Make You Do

GUI-only management doesn't
survive contact with scale.

Most traditional appliances — and even some "API-enabled" ones — treat the API as an afterthought bolted onto a GUI-first product. The difference shows up the moment you try to manage more than one box.

Common patternWhere it breaks down
GUI-first appliances with a thin APIThe API often only covers a fraction of what the GUI can do — bulk rule import, DDoS tuning, or export may be GUI-exclusive, forcing a human back into the loop for exactly the operations automation needs most
Vendor-specific config languagesProprietary CLI syntax (not REST, not standard SQL/YAML) means every integration is custom-built and brittle to firmware upgrades — there's no ecosystem of existing tools to reuse
No dry-run / simulation capabilityA scripted rule push either works or breaks production — there's no way to test a change's real impact before it's live, so teams either skip automation for risky changes or accept the blast radius
Change tracking bolted on after the factAudit logs that only capture GUI actions miss automated changes entirely — the exact changes you'd most want traced during an incident review are invisible
What Happens Without a Real API

Manual firewall management
doesn't fail loudly — it fails slowly.

You need to block an IP during an active incident
Without an API, someone has to log into a GUI, find the right screen, and click through a form — adding minutes to a response that should take seconds, while the attack traffic keeps flowing.
You manage rules across multiple gateways
Manual changes replicated by hand across sites inevitably drift — one gateway ends up with a rule the others don't, and nobody notices until an audit or an incident exposes the inconsistency.
A compliance auditor asks who approved a rule change
Without rule changes flowing through Git/CI, the answer is "someone, sometime, in the GUI" — with no pull-request trail, no reviewer, and no diff to show. That's a finding, not an answer.
You push a config change and it's wrong
Without a simulate-before-apply step, you find out it's wrong from a downed service or angry users — not from a pre-merge check that would have caught it in seconds.
Why CloudArmour

The API isn't an add-on —
it's the primary interface.

Full Feature Parity

Every capability the UI exposes — rule CRUD, DDoS config, threat intel, audit queries, compliance export — is reachable through the same REST API, not a stripped-down subset.

Simulation Built In, Not Bolted On

Policy trace simulation was part of the core design, not a feature added after customers asked for safer automation — see Automate Your Firewall for the guardrail patterns this enables.

Standard Auth, Standard Tools

API keys, JWT bearer tokens, and RBAC work with any HTTP client or CI runner you already use — no proprietary SDK, no vendor lock-in to a specific automation platform.

Every Automated Change Is Audited

API-driven mutations are captured by the same immutable audit log as manual changes, with the calling identity attached — automation doesn't create a blind spot in your audit trail.

Get started

Put the firewall in your pipeline.