A customer just asked how you handle their data. Have the answer.
You're moving fast — but one security questionnaire or a leaked NIK can stall the deal. Anoman gives every AI call prompt-injection defense and PII masking on a Jakarta-hosted gateway — with a receipt you can show your customer. Start free.
You shipped the demo. Then came the questionnaire.
- You closed the demo — then the security questionnaire landed, and you didn't have an answer. The deal waits.
- You're one prompt injection or leaked NIK/email away from being the breach story — on a 72-hour UU PDP clock with up to 2% of revenue at stake.
- Every foreign-model call sends your customer's data abroad. No residency, no audit trail, no proof.
- You don't have the time — or a security team — to build guardrails, PII masking, and audit logging yourself.
Anoman ships all of it on every call — so you can keep moving fast and still answer “is it safe?” with proof.
From npm install to guarded in minutes
Three steps to production
Point at Anoman
Set OPENAI_BASE_URL=https://api.anoman.io/v1 — your existing OpenAI or Anthropic SDK client works unchanged. Two lines.
Every call is guarded
Prompt-injection defense, PII masking, content moderation, and policy run on every request — before it reaches any model. Nothing to wire up.
Show the receipt
Each response returns an _anoman evidence block (region, guardrail results, routing, cost) — your proof for the next security questionnaire.
# Your existing OpenAI code — change ONE line:
export OPENAI_BASE_URL=https://api.anoman.io/v1
export OPENAI_API_KEY=anm-sk-...278,111 AI calls · 657.4M tokens since May 2026
The plan that fits a startup
Free to start — guarded from your very first call
- Up to 2M tokens/day, best-effort
- Full guardrails on every call
- $3 credit + premium-model trial
Start free, no credit card. Cancel anytime — drop-in and drop-out with two lines of config. No lock-in.
Startup questions, answered
The things a founder actually asks before shipping.
How much does it cost to start?
Free — start on the Free tier, no card required: guarded calls on budget + mid models every day, plus a $3 signup credit. When you need more, top up a prepaid balance in Rupiah (raw provider cost, 0% token markup — a small platform fee applies at top-up) or move to a day/week Pass or Pro. No upfront commitment.
Is there any lock-in?
No. Anoman is OpenAI-compatible — you changed two lines to come in and can change them back to leave. No proprietary SDK.
Is my data safe from Anoman itself?
Anoman never trains on your data, and guarded models' providers never train on it either; the may-train community lane is opt-in. Cache is partitioned per customer, API keys are bcrypt-hashed, and PII is masked on every request before it reaches a model.
Where does my data actually go?
Account data and logs are stored in Jakarta (subprocessors listed on the privacy page), where the gateway runs. A request only crosses a border if the model you choose is processed elsewhere — shown on that model's page. No silent cross-border transfers.
Why trust a newer, Indonesia-based vendor?
Every response returns a verifiable receipt of exactly what we did, we run a public status page and the Founding 50 program, and Anoman is built by security experts with more than 15 years in enterprise cybersecurity.
Your customers are already asking. Be ready.
Ship AI you can prove is safe — guarded, logged, and routed through our Jakarta gateway on every call.