No marketing language. No weasel words. Each commitment below is enforced by code or contract — if we ever break one, write us at [email protected] and we'll make it right.
Subscription cancel works in one form at /billing/cancel — no retention call, no sales rep, no hidden steps. We do show a 'stay one more month on us' offer (you can ignore it), but the cancel button is right there on the same page. If you change your mind later, we'll be here when you come back.
Cancel mid-cycle and get a pro-rated refund for the unused portion. No fine print, no admin fee. Refunds settle on Stripe within 5-10 business days back to the original payment method.
Set a monthly spend cap from your account dashboard (or via the /billing/auto-recharge?email=... endpoint) and it's enforced by shared/usage_governor.py — we technically cannot bill you over your cap. If we hit it, your sub-app pauses and emails you. No credit card surprise.
GDPR Article 7(3) says withdrawal of consent must be as easy as giving it. We comply: cancel + downgrade + data export are all at the same depth as signup, behind one click.
Sign in to your account and hit /api/account/export?email=<your-email> to download a JSON archive of every prompt, output, audit event, and billing record we have. No support ticket, no waiting period, no sales objection.
Every AI call Manera makes is logged with timestamp + flagship + sub-app + prompt-cache hit/miss + token cost. Available on enterprise SKUs as a SIEM-exportable feed (per-output SHA-256 hashing is on the Sovereign-tier roadmap — ask if you need it as a contract clause).
If we change pricing for a SKU you're subscribed to, you get 60 calendar days of notice via email + in-app banner. You can lock in the old price for the rest of your annual term, or cancel pro-rated.
Not to data brokers, not to lookalike-modeling services, not to ad networks. Our revenue model is subscription + usage; we have zero incentive to monetize your data.
Anthropic's API terms (which govern our inference) bar training on API customer inputs by default — see Anthropic's Commercial Terms + Trust Center. Your prompts hit the inference endpoint and disappear (subject to ephemeral logs for 30-day debugging retention only). Documented in /privacy.
What we ship is at /changelog. What we plan is at /roadmap. If a feature you depend on is being deprecated, it appears on the roadmap with at least 90 days notice before deprecation.
Email [email protected]. We respond within one business day. If we agree we broke a commitment, we issue a credit (minimum: one month free of your highest tier) and post a public retro at /changelog. We'd rather lose money than reputation.