Behavioral Runtime Protection for PHP

Stop attacks before they
reach your code.

SentraShield is a lightweight, behavior-learning WAF/IDS that runs via auto_prepend_file. It learns your application's legitimate request patterns - then blocks anything that structurally deviates. No agents. No root access. No code changes.

Shared hosting compatible Zero code changes required Multi-domain licensing
Live Protection Status
Integration
1 line
Blocked Today
247
Routes Trusted
94%
Overhead
<0.5ms
Monitor
Simulate
● Protect
Compatible with →
Apache mod_php PHP-FPM 7.4/8.x cPanel / Plesk LiteSpeed LSAPI Shared hosting VPS / Dedicated
Capabilities

Everything you need. Nothing you don't.

Designed for LAMP stacks, agencies, and hosting environments. SentraShield protects your PHP applications without touching your application code.

🧠

Behavioral Learning Engine

Automatically learns your application's legitimate request structure - parameter types, route schemas, and normalized input patterns. No manual rule creation.

🛡️

Monitor → Simulate → Protect

Start safely in Monitor mode, observe potential blocks in Simulate, then switch to full protection. The lifecycle is designed to prevent false positives.

Runs Before Your Code

Integrates via auto_prepend_file. Every request is inspected before your application executes. Pure PHP - no Zend extension, no C agent, no root access.

🌐

Multi-Domain Licensing

Manage multiple PHP sites under a single license key. Domain registration is automatic, enforced server-side. Ideal for agencies and hosting providers.

📊

Security Dashboard

A built-in admin panel shows baseline quality, blocked threats, parameter schemas, route coverage, and NDJSON security logs - all without a separate server.

🗃️

SIEM-Ready Logging

All security events are written as structured NDJSON. Feed directly into your SIEM pipeline or analyze locally. Every request is a structured, queryable log entry.

🔒

Policy Freeze / Lock

Once your baseline is stable, freeze it into a policy. Learning stops, the schema is locked, and only known-good request patterns are allowed through.

📦

Multi-App Isolation

Each application gets its own isolated baseline, schema, policy, and log directory under .sentrashield/apps/. Multiple sites on one server stay fully separated.

🚀

Low Overhead by Design

Sub-millisecond inspection on typical requests. License validation is cached for 24 hours. Async refresh happens after response - zero latency impact for visitors.

Lifecycle

Learn. Freeze. Protect.

SentraShield follows a deterministic three-phase lifecycle that gives you full control over when and how protection is applied.

01

Install & Learn

Drop the .sentrashield/ folder on your server and set auto_prepend_file. SentraShield immediately begins learning your application's legitimate traffic patterns - routes, parameter types, schemas.

baseline.json schema.json Monitor Mode
02

Simulate & Verify

Switch to Simulate mode and run the built-in test suite. SentraShield logs what it would have blocked without touching real traffic. Tune the baseline until false positives reach zero.

Simulate Mode Test Suite policy.json
03

Freeze & Protect

Freeze the schema into a policy and enable Protect mode. Requests that structurally deviate from your learned baseline are blocked in real-time - before your application code ever executes.

Protect Mode Real-time Blocking NDJSON Logs

Ready to protect your PHP sites?

Start in Monitor mode - it's free during the 30-day trial, no disruption to live traffic.

Security Model

Positive security. Not pattern matching.

Traditional WAFs ask: "Is this a known attack?" SentraShield asks: "Does this request match the learned, legitimate behavior of my application?" That's a fundamentally stronger model.

Attacks Detected (without signatures)

  • SQL Injection
  • Path Traversal / LFI
  • Server-Side Request Forgery (SSRF)
  • PHP Object Injection
  • Template Injection (SSTI)
  • Encoding bypass variants
  • Manipulated parameter structures
  • Parameter pollution attacks
  • Schema & presence violations

Technical Strengths

  • No C agent - pure PHP
  • No root access required
  • No Zend extension risk
  • Shared hosting compatible
  • Deterministic policy model
  • NDJSON / SIEM-ready output
  • CLI lifecycle management
  • Sensitive fields auto-redacted in logs
  • No Authorization headers or cookies ever stored
  • GDPR-aware logging by design
Not: "Is this a known attack?"

But: "Does this match legitimate behavior?"

SentraShield works with behavioral allowlisting - not signature blocking. After learning your application's structure, it fixes that model as a policy. Anything outside the policy is structurally anomalous and blocked.

This is the classic positive security model: learn legitimate structure → fix it → block structural deviations. This approach catches zero-day attacks and novel payloads that signature-based systems miss entirely.

Operating Modes

Full control at every stage.

Three clearly defined modes let you adopt SentraShield at your own pace - with zero risk to production traffic.

👁️

Monitor

SentraShield observes all traffic, builds the behavioral baseline and parameter schema. Nothing is blocked. Safe for immediate deployment on any live site.

LEARN + OBSERVE
🔬

Simulate

Blocking rules are active but attacks are logged, not blocked. Run the test suite and verify detection coverage. Tune until zero false positives before enabling protect.

LOG, DON'T BLOCK
🛡️

Protect

Full enforcement. Requests that deviate from the learned policy are blocked in real-time before your application executes. Automatic downgrade to Monitor if license expires.

REAL-TIME BLOCKING
Dashboard

Built-in security intelligence.

The SentraShield dashboard gives you full visibility into your protection status, threat events, baseline quality, and schema coverage - no external tools required.

View Details
Overview & License Status
View Details
Baseline & Route Coverage
View Details
Security Event Log
View Details
Schema & Policy Routes
View Details
Domain Management
View Details
Test Suite Results

Click any screenshot to enlarge · Real dashboard screenshots coming with your trial

Technical Transparency

What SentraShield is - and what it isn't.

We believe in technical honesty. SentraShield is a specialized tool that does its job exceptionally well. Here's exactly where it fits.

SentraShield is…

  • A lightweight PHP runtime engine for behavioral rules
  • A lifecycle-driven policy enforcement system
  • A structured telemetry and threat logging layer
  • Optimized for LAMP, cPanel, and shared hosting
  • Suitable for legacy PHP applications
  • Developer and DevSecOps friendly

SentraShield is not…

  • A classic network WAF (no deep packet inspection)
  • A full RASP with C-level sink hooking
  • An IAST for code-level vulnerability discovery
  • A DB query parser (validates before execution)
  • An APM-integrated security platform
  • A replacement for secure coding practices
Pricing

Simple, domain-based pricing.

Every plan includes a 30-day free trial. Upgrade, downgrade, or cancel anytime.

Need more domains or a custom SLA?

Enterprise plans include dedicated support, custom domain limits, and a service level agreement.

FAQ

Common questions.

Everything you need to know before getting started.

Does it require changes to my application code?

No. SentraShield integrates entirely via auto_prepend_file in your .htaccess or php.ini. Your application code is completely unchanged.

Will it break my live site?

No. You start in Monitor mode which only observes traffic - nothing is blocked. You only enable blocking after verifying zero false positives in Simulate mode.

Does it work on shared hosting?

Yes. SentraShield is specifically designed for shared hosting. It requires only PHP 7.4/8.x, no root access, no server extensions, and no compiled modules.

How does multi-domain licensing work?

Each plan includes a domain limit. When you enter your license key on a new domain, it's automatically registered. The server enforces the limit - no manual configuration needed.

What happens when my license expires?

SentraShield automatically downgrades to Monitor mode - your site stays online, traffic continues to flow, but active blocking is paused until you renew.

Can I try it before committing?

Yes. Every plan includes a 30-day free trial. Start in Monitor mode with zero risk, verify it works for your application, then decide whether to continue.

Does SentraShield log sensitive user data like passwords or tokens?

No. SentraShield is built with privacy-aware logging from the ground up. Sensitive form fields — passwords, tokens, API keys, CVV numbers — are automatically detected by parameter name and replaced with ***REDACTED*** in all log files before anything is written to disk. Bearer tokens and API keys in URLs or Referer headers are stripped before traffic logs are written. HTTP Authorization headers and cookies are never read or stored. The behavioral baseline and schema store only abstract value types (:int, :str, :uuid) — never actual user input. All logs remain exclusively on your server.

How does learning work? Can it be fooled?

SentraShield learns semantic structure, not concrete values. Values are normalized (integers become :int, UUIDs become :uuid) before storage - preventing overfitting and value-binding attacks.

Is it compatible with WordPress, Joomla, Laravel, etc.?

Yes. SentraShield works with any PHP application. The learning phase adapts to your specific application's request patterns regardless of the framework.

Contact

Get in touch.

Have questions, need an Enterprise quote, or want to discuss a custom integration? We respond quickly.