Download the PHP package nemesislabs/sentinel without Composer
On this page you can find all versions of the php package nemesislabs/sentinel. It is possible to download/install these versions without Composer. Possible dependencies are resolved automatically.
Download nemesislabs/sentinel
More information about nemesislabs/sentinel
Files in nemesislabs/sentinel
Package sentinel
Short Description Nemesis Shield - Sentinel SDK for PHP & Laravel. Positive-security runtime protection: learns your app's normal behavior and blocks off-baseline requests (auth bypass, path traversal, scanners, unusual methods). Fail-open, privacy-preserving. Works on Laravel 6+ / PHP 7.2+.
License MIT
Homepage https://nemesislabs.xyz
Informations about the package sentinel
Nemesis Shield - PHP
Native PHP SDK for Nemesis Shield. Learns your app's normal behavior; in enforce mode BLOCKS off-baseline requests (auth bypass, path traversal, scanners, unusual methods) before your app runs. PHP is stateless per-request, so the compiled policy is cached to a temp file with a short TTL and refreshed on demand. Positive-security, fail-open.
Raw / any framework
Laravel (Composer, 6+ / PHP 7.2+) - zero-config via package auto-discovery:
NemesisShieldServiceProvider prepends the middleware to the global stack automatically - no
Kernel.php edit. Full guide + manual (no-Composer) fallback + Namecheap/shared-hosting notes:
examples/laravel/. (No Composer? drop NemesisShield.php +
NemesisShieldMiddleware.php in and register the middleware yourself - see the example.)
Symfony - register it as a PSR-15 middleware / kernel subscriber wrapping guard/observe.
Observe (default) → learn & approve in the console → flip to enforce → off-baseline requests get
403 blocked_by_nemesis_shield. Verified end-to-end (learn → enforce → attack) on raw PHP: legit
passes (200); attacks blocked (403).
LLM Guard (OWASP LLM Top 10)
The same HashLR ML classifier every Nemesis Shield SDK ships - catches obfuscated prompt injection signature rules miss, scored identically in every language.
Regex first, then ML. Blocks at ≥ 0.85 (high), flags at ≥ 0.45.
Full coverage & safe-unlock
Mount it first / outermost so every route is inspected (not just API routes - attackers hit any path):
What's inspected (privacy-preserving): method + normalized route + query-param structure (names + kinds, never values) + auth flag + status. An off-baseline route, param structure, method, or auth state is blocked in enforce mode. Path-traversal segments normalize to {traversal}.
Safe-unlock (break-glass): the login/auth path is never blocked, so a still-learning baseline can't lock you out. Defaults: /login /signin /sign-in /auth /oauth /session /wp-login.php /wp-admin. Override:
Verify coverage - in observe mode, hit a normal route, a param, and a scanner path, then confirm all three appear in the console (Activity / Behaviors):