Everything about buying, deploying, and operating a managed agent. If something's missing, contact support — we'll answer and add it here.
Getting started
How purchasing and provisioning works, end to end.
How do I buy an agent?
Open any agent page and choose your delivery option — one-off build or managed service. Check out with USDT or USDC (stablecoin). Your order workspace is created immediately, and we start on your deployment once you submit the required setup details.
Do I need to create an account before checkout?
No. Your workspace is created automatically when your payment confirms. Access is granted via a secure magic-link email — no password, no account form to fill in mid-purchase.
How long does deployment actually take?
The standard SLA is 48 hours from the moment we have everything we need from you — your scoped credentials, confirmed integration details, and any platform access. The order workspace shows exactly what's outstanding so there's no guessing. Some agents carry a longer SLA; it's stated on each agent page.
What is the difference between a one-off build and a managed service?
A one-off build is a single payment. We configure, test, and hand over the deployment with full documentation — after that, it runs on your infrastructure and API costs. A managed service includes ongoing hosting, monitoring, maintenance, and small updates under the agreed monthly fee. You never have to deal with infrastructure yourself.
Can I request something not in the catalogue?
Yes. Use the custom quote request form to describe the outcome you need. We'll review the scope, ask any clarifying questions, and come back with a proposal and price before any work begins.
Setup & credentials
What we need from you, and what we will never ask for.
What credentials will you ask for?
Only scoped, revocable tokens for the specific integration in question. For example: a Discord bot token with defined permissions, a read-only on-chain RPC endpoint, or a webhook destination URL. We specify exactly what's needed per agent, and it's always the minimum required.
Will you ever ask for a private key or seed phrase?
Never. We do not ask for, store, or accept private keys, seed phrases, raw passwords to third-party services, or admin credentials at any point in the process. If a message claims to be from us and asks for any of these, it is not from us.
Can I revoke access after deployment?
Yes. Every credential you supply is scoped and revocable. The handover runbook explains exactly how to rotate or revoke each one. Revoking access does not void your 30-day coverage — we can work with fresh credentials if needed.
What if I don't have the required credentials ready at checkout?
That's fine. You can purchase and set up your workspace first, then submit credentials when you're ready. The 48-hour deployment clock starts when we have everything — not when you pay.
Orders & workspace
Tracking progress, milestones and messages.
How do I know what is happening with my order?
Your order workspace is the source of truth. It shows the current status, milestone timeline, messages from your engineer, anything we need from you, and the delivery details at handover. You never need to contact support just to get a status update.
What are the order statuses?
Payment Confirmed → Setup Details Required → Ready for Deployment → In Progress → Testing → Awaiting Your Review → Completed. Managed service orders continue to Ongoing Service after completion. You may also see Waiting on Customer (we need something from you) or Blocked (an external issue is pausing work).
How will I be notified of changes?
Status changes, milestone updates, and engineer messages all trigger in-app notifications and an email to the address on your order. You don't have to check the workspace constantly — we'll tell you when something needs your attention.
Can I send messages to my engineer?
Yes. The workspace has a message thread that goes directly to the engineer assigned to your order. This is the fastest way to ask questions or provide additional context during deployment.
Billing & refunds
Pricing, recurring costs, and how to cancel.
What am I charged at checkout?
For a one-off build: the full build fee, once. For a managed service: the setup fee at checkout, then a monthly recurring fee starting after handover. The exact amounts are shown on the agent page and at checkout before you pay — nothing is charged that isn't disclosed upfront.
How do recurring monthly charges work?
After handover of a managed service, you receive a monthly invoice via email with a crypto wallet address and exact amount. You send the payment — it doesn't auto-charge. If a payment is late, we'll reach out before pausing service. Recurring terms are always shown on the agent page before you buy.
Can I cancel a managed service?
Yes. Contact support or message your engineer through the workspace to request cancellation. There's no minimum term unless separately agreed. Cancellations are processed at the end of the current paid period — you won't be charged for a month you've already paid.
What is the refund policy?
If we haven't started your deployment yet and you want to cancel, you're entitled to a full refund. Once deployment has begun, refunds are prorated based on work completed. The full refund policy is at /policies/refunds.
Can I upgrade from a one-off build to a managed service?
Yes. If you purchased a one-off build and want to move to managed hosting and maintenance, contact support. We'll scope the transition and set up a managed service agreement — you don't have to start over.
Security & architecture
How agents run and how your data is treated.
Where do agents run?
Each deployment runs in a dedicated isolated runtime with a least-privilege token model. Your agent's operational data and conversations stay within its own environment — not shared with other customers' deployments.
What happens if an agent fails or behaves unexpectedly?
Every deployment ships with failure-scenario fallbacks and alerting. During the 30-day coverage window, your engineer is notified automatically and will investigate. After that window, managed service customers have ongoing monitoring included. One-off builds can arrange support separately.
Do I get the source configuration and runbooks?
Yes — always. Handover includes the complete deployment runbook: what the agent does, what each component is, how to restart it, how to rotate credentials, and what to do in specific failure scenarios. You own the deployment and can operate it independently.
Who can see my order data and credentials?
Only the engineer assigned to your order and authorised platform staff. Order data is protected by row-level security in the database — other customers cannot access your order, messages, or setup details, even in theory.
Integrations & platforms
Supported platforms and what to prepare.
Which platforms are supported?
It varies by agent. Across the catalogue: Discord, Telegram, Slack, GitHub, Linear, Zendesk, PagerDuty, Notion, Snapshot, Discourse, and major EVM and Solana networks. Each agent page lists its supported integrations.
Can you integrate with a platform not listed?
Often yes. Contact support with the platform and the agent you're considering — we'll confirm feasibility before you pay. If we can do it well, we'll say so. If we can't, we'll tell you that too.
Do I need any technical knowledge to purchase?
No. You need to be able to generate API keys or bot tokens for the platforms your agent will use — most platforms have a clear guide for this. Your engineer will tell you exactly what's needed and walk you through anything unclear.
Still stuck? Our team answers during support hours — no sales call needed.