What is an LLM vulnerability scanner?

An LLM vulnerability scanner sends batteries of adversarial probes — jailbreaks, prompt injections, PII extraction attempts, harmful-content requests — at a target LLM and grades the responses. The output is a vulnerability + optimization report with per-probe evidence, severity, and a prioritized fix list.

Which LLMs can I scan with FilterPrompt?

OpenAI, Anthropic, Google Gemini, Azure OpenAI, plus any OpenAI-compatible endpoint — Ollama, Groq, Mistral, Together AI, OpenRouter, Perplexity, Hugging Face, vLLM, or your own custom endpoint. Bring your own keys per tenant.

What kinds of vulnerabilities does FilterPrompt test for?

Jailbreaks (DAN, role hijack, translation smuggling), direct and indirect prompt injection, system-prompt extraction, harmful-content compliance, PII / secret leakage, bias & fairness, RAG poisoning, agent/tool abuse, output quality, and robustness — categories map to the OWASP LLM Top 10.

How are probes graded?

Each probe declares an evaluator: regex match, refusal-check, contains-check, or an AI judge (Gemini 3 Flash). Pass/fail comes with severity, category, the exact prompt sent, the model's full response, and the evaluator's reason — fully auditable.

How much does a scan cost?

1 credit per probe executed. New accounts get 1 welcome credit on signup. Pay-as-you-go credit packs after that — credits never expire. Connecting LLMs and creating tenants is free.

What is a multi-tenant AI gateway?

A multi-tenant AI gateway is a single proxy that handles LLM traffic for many isolated tenants — applying per-tenant rate limits, quotas, model allowlists, IP rules, and audit logs while routing to each tenant's own provider credentials.

Does FilterPrompt store provider API keys?

Each tenant brings their own provider keys (OpenAI, Anthropic, Gemini, Azure OpenAI, Bedrock). Keys are encrypted at rest and never logged. FilterPrompt never holds shared keys, so cost and compliance attribution stay with the tenant.

Multi-Tenant AI Gateway with Per-Tenant Quotas

Run a single secure gateway for all tenants. Per-tenant rate limits, quotas, model allowlists, IP rules, and isolated audit logs — across OpenAI, Anthropic, and Gemini.

Tenant isolation

Every request carries a tenant identifier resolved from the x-firewall-key header. Logs, quotas, rules, and provider credentials are partitioned per tenant — no cross-tenant data exposure even in shared deployments.

Controls per tenant

Rate limits — requests/minute, tokens/minute, concurrent sessions
Monthly quotas — hard caps with overage alerts
Model allowlists — restrict which providers and models a tenant can call
IP allow/blocklists — global, per-tenant, and auto-ban on abuse
Custom rules — bring your own regex/keyword detectors

Provider credentials

Each tenant brings their own provider keys (OpenAI, Anthropic, Gemini, Azure). FilterPrompt never holds shared keys, which means costs and compliance attribution stay with the tenant.

AI firewall overview
Designing a multi-tenant AI gateway
Firewall vs gateway

Multi-Tenant AI Gateway with Per-Tenant Quotas

Tenant isolation

Controls per tenant

Provider credentials

Related