Managed platform
The entire platform — prompt-injection firewall, de-obfuscation engine, output guardrails, contract analyzer, metering — runs as a managed cloud service. No setup, no maintenance. Onboarding opens soon.
Pricing · status
Hosted tiers are
Hosted tiers are
being finalized.
We're not publishing prices for plans you can't buy yet — that's how trust dies. What we can promise today: the full platform runs as a managed service, and onboarding opens soon.
Hosted API tiers — coming soon
A managed gateway with usage-based tiers is in the works: same engine, zero infrastructure on your side, SLA targets, and EU AI Act reporting built in. When the meters run live, the prices get published here — from the metering service, not from marketing.
Why "coming soon" instead of a price table?
Publishing tiers requires three things working end-to-end: a hosted gateway, metered enforcement, and checkout. Two of three exist in code today; none are exposed publicly yet. Every claim on this site has a receipt — pricing will too, the day it's real.
Meanwhile
Free things you can use today.
The engine
Both security directions — prompt screening and response guardrails — plus the de-obfuscation layer. Full behavior, not a crippled community edition.
SDKs & proxy
Python and JavaScript SDKs, plus the universal auth proxy for Ollama, vLLM, LocalAI, and OpenAI-compatible endpoints.
The receipts
Benchmarks, chaos results, test counts, and audit status — all published with sources on the proof page, including the unflattering rows.
Don't wait for pricing.
Watch it work.
The demo runs real detection rules in your browser — no signup, no gateway required.