Free Browser Tools

AI Security Toolkit

Free, in-browser tools to scan, protect, monitor, and comply. Need more? The hosted API adds an enhanced engine and saved history.

Live now · free

Analyze (Unified Report)

One prompt, full report: prompt-injection + secrets/PII + system-prompt audit + an optional LLM second opinion. Runs in your browser; adds an enhanced hosted engine and API. Replaces the standalone scanner, PII, and audit tools.

Launch Analyze

Agent Guard (Tool-Call & MCP)

Audit your agent's tool/function definitions — OpenAI, Anthropic, or MCP — for excessive agency, code-execution and SSRF sinks, secret access, and data-exfiltration chains. Mapped to the OWASP 2025 LLM Top-10. Runs in your browser; the hosted engine adds cross-tool analysis.

Launch Agent Guard

LLM Gateway (Secure Proxy)

A firewall in front of any model: it scans the prompt, forwards it to the provider/model you pick (OpenAI, Anthropic, xAI, Groq — BYOK), then scans the response — masking leaked secrets/PII and blocking prompt-injection inline. Choose flag or block policy. Deterministic engines stay authoritative.

Launch Gateway

Prompt Firewall

A guardrail that decides whether text should be allowed, flagged, or blocked before it reaches an LLM — combining injection, secret, and PII detection under a policy you control.

Launch Firewall

Red-Team Toolkit

Generate adversarial test prompts (jailbreaks, instruction overrides, prompt leaks, data exfiltration) to harden your own LLM app, with a response-scoring checklist.

Launch Toolkit

Compliance Self-Assessment

Score your AI/LLM deployment against controls from the OWASP LLM Top 10 and EU AI Act / NIST AI RMF / ISO 42001, and get a graded report with prioritized gaps.

Start Assessment

Model Integrity Checker

Hash a model, dataset, or config file with SHA-2 and verify it against the publisher's checksum — plus an AI supply-chain checklist. Files never leave your browser.

Launch Checker

LLM Observability

Monitor your LLM app for injection attempts, blocks, and anomalies. Send security events via API key and see them aggregated on your account — authenticated scans show up automatically.

Open Observability

Quick Start

Terminal

$ npm install @opensecureai/scanner

$ npx -y @opensecureai/scanner scan ./prompts/*.txt

$ npx -y @opensecureai/scanner scan prompt.txt --fail-on high