SOFTWARE API

EvrGuard API

Secure any AI agent in 3 lines of code.

Mathematical verification governing every AI action. Cryptographic proof on every decision.

Real-time interception of a prompt injection attack

Works with any AI agent.
Claude CodeClaude DesktopOpenClawLangChainLangGraphAutoGenCrewAIAny Python Agent
How It Works

Three steps to deterministic safety

STEP 1

CONNECT

Wrap your agent with the SDK, MCP server, or proxy. Three lines of code. Any framework.

STEP 2

VERIFY

Every action is mathematically verified through neuro-symbolic gradient steering. Cryptographic proof on every decision.

STEP 3

ENFORCE

Unsafe actions are blocked. Safe actions proceed. Hardware enforcement available for critical systems.

main.py
1
2
3
4
5
6
7
8
from evrguard import EvrmindGuard
guard = EvrmindGuard(api_url="https://api.evrmind.io")
@guard.wrap_tool
def execute_query(sql):
return db.execute(sql)

Every tool call passes through the safety engine. PASS, STEERED, or BLOCKED. Proof hash on every decision.

Intelligent Safety

Safety That Steers, Not Just Blocks.

When an action violates a rule, EvrGuard steers the agent toward the nearest safe alternative. Computed through gradient descent on a constitutional manifold. The agent adjusts and continues. No stuck states. No manual intervention needed.

1
Agent Proposes
"Export customer records with home addresses to external partner"
EvrGuard Verdict: STEERED
2
Safe Alternative
"Export anonymised customer analytics to internal reporting system"
Agent continues working
Integration Methods

Three ways to connect

Python SDK

Wrap any function. Works with LangChain, LangGraph, AutoGen, CrewAI, or custom agents.

pip install evrguardView on GitHub

MCP Server

Zero code integration for Claude Code, Claude Desktop, and OpenClaw. Add one config line. Every tool call is verified automatically.

pip install evrguardSetup guide

HTTPS Proxy

Works with any language, any application. Set your proxy variable. All HTTP traffic passes through EvrGuard.

pip install evrguardSetup guide
Features

Everything you need for agentic safety

Triple-Layer Detection

Every action passes through all three detection layers automatically. Pattern matching runs first. The engine safety net catches what patterns miss. Intent analysis catches the rest. One product. Three layers. No configuration needed.

Neuro-Symbolic Steering

Not just block or allow. Mathematical gradient descent steers unsafe actions toward safe alternatives. 60 iterations on a constitutional manifold.

Cryptographic Proofs

HMAC-SHA256 signed verdicts. Merkle root anchoring. Tamper-evident audit trail. Regulators can independently verify every decision.

10 Industry Constitutions

Pre-built safety rules for enterprise, finance, healthcare, legal, DevOps, retail, education, insurance, government, and critical infrastructure. Deploy in minutes.

OWASP Agentic Top 10

Full compliance pack covering prompt injection, tool abuse, privilege escalation, data exfiltration, and all 10 OWASP agentic security categories.

Monitor Mode

Run alongside your agents without blocking anything. See what would be caught. Build confidence in your rules. Switch to enforce when ready.

Regulatory Compliance

Pre-Built Regulatory Compliance

Select the frameworks that apply to your organisation. EvrGuard enforces them automatically.

compliance.evr
# EvrGuard Compliance Rules (2 frameworks selected)
REQUIRE: REQUIRE_HUMAN_OVERSIGHT_FOR_AUTOMATED_DECISIONS [strength=75.0]// EU AI Act
REQUIRE: NO_PII_EXPORT [strength=80.0]// GDPR

Plus: DORA, NIS2, CCPA, COPPA, FERPA, PCI-DSS, and industry-specific constitutions for 10 sectors.

*Compliance packs provide technical enforcement infrastructure and do not constitute legal advice. Review rules with legal counsel for your specific use case.

Constitutional Compiler

Write your own rules. Or choose from our library. Enforce everywhere.

Define your safety rules in a simple, human-readable constitution. Our compiler translates them to both software verification and hardware enforcement.

my-constitution.evr

Software

EvrGuard API

Neuro-symbolic verification in <1ms

Hardware

Gate-level enforcement in 80ns

Pricing

Simple, transparent pricing

Free

£0/month
  • 1,000 verifications per month
  • Choose from 10 templates
  • Documentation and guides
Get Free API Key

Starter

£69/month
  • 10,000 verifications
  • Custom constitutions
  • Webhook alerts
  • Email support
Subscribe
Most Popular

Professional

£279/month
  • 100,000 verifications
  • All compliance packs
  • PDF reports
  • Compliance dashboard
  • Priority support
Subscribe

Enterprise

Custom
  • Unlimited verifications
  • Dedicated instance
  • SLA
  • Hardware upgrade path
  • Auditor portal
The Core Difference

Why Mathematical Verification Changes Everything

Prompt Guardrails

Asks the AI to behave. Defeated by jailbreaks.

AI Safety Layers

Uses AI to check AI. Same failure modes.

EvrGuard

Pure mathematics. Six patented algorithms. Cryptographic proof.

97.3% hostile catch rate. And when you need absolute certainty, the same rules compile to silicon.

Ready to secure your AI agents?

Start with 1,000 free verifications per month. No credit card required.

Get Started Free