Build your AI agent, see the price
Pick the knowledge, the tools, and the guardrails your agent needs. The itemized scope and the fixed price update as you choose; every build starts from the autonomous agent base.
Your starting build
Recommended for you
Most clients start here. Every module is priced; change anything below and the total follows.
≈$14,990plus VAT as required by lawbilled as ₪45,000
- Autonomous AI agent (base)Included in every build≈$8,330
- Knowledge base: Retrieval over your documents (RAG)Answers are grounded in your document library.+≈$1,500
- Tool integrationOne connected system the agent can act in.+≈$840
- Memory model: Long-term memoryNobody repeats themselves across conversations.+≈$1,000
- Guardrail tier: StrictAn allowed-topics list and a no-hallucination policy.+≈$1,000
- Conversation tracingEvery tool call and decision visible, so you can debug a bad answer.+≈$1,000
- Analytics dashboardVolume, resolution and handoff rates against your goals.+≈$840
- Cost monitoring and budgetsModel spend tracked with budget alerts.+≈$500
Start from a ready package
Ready-made bundles for common needs. Pick one, then adjust anything.
Autonomous AI agent (base)
≈$8,330
Included in every build and non-negotiable: no module removes the grounding, safety, or oversight baseline.
How your agent is built
Not a chatbot bolted onto a prompt. A grounded, tool-using system engineered for safety and ownership: it reasons over your knowledge, acts through real integrations, and stays observable in production.
The agent brain
- A working agent grounded in a starter knowledge base you provide: it answers from your content, not the open internet
- Tool use over one system you already run: the agent reads or acts through a real integration, not just chats
- Within-conversation memory and deterministic intake, so a multi-step request holds together
- Runs on the right model for the job, chosen for your case, with data minimized to what the model needs
Safety and reliability spine
- Baseline guardrails: answers fenced to your knowledge, prompt-injection defenses, and refusal on out-of-scope requests
- A human handoff with full context whenever the agent is unsure or a person is asked for
- Full run logging with PII minimized, so you can see what the agent did and why
- Deployed to managed cloud on your own domain, verified against your real scenarios before launch
Ownership and delivery
- The full source is yours, in your own repository, with no lock-in to us
- Documentation and runbooks your team can actually read, so you understand how the agent works
- A handover walkthrough with your team, so nobody is left guessing
- A 30-day warranty after go-live; ongoing care is optional and monthly, never a lock-in
We run agents on our own operations first
The same grounded brain, guardrails, evaluation harness, and human-in-the-loop oversight run the internal agents we rely on. We do not ship an autonomy pattern we would not trust ourselves.
Build it your way
Each capability shows its fixed price. Where the real work is curating your knowledge or learning a system we have not touched before, the discovery share is priced visibly.
What the agent knows, and how it stays current. The base grounds answers in a starter set; this is the real retrieval brain over your content.
What the agent can DO, not just answer. Each tool lets it act in a real system, with auth and safe-write guards.
Where the agent lives. The base ships a web chat widget; add the channels your users and staff actually use, all sharing one brain.
How much the agent remembers. The base holds the current conversation; go further with long-term memory and per-user personalization.
How the agent thinks. The base is a single capable agent with tool use; deepen it to multi-step planning or a coordinated multi-agent system.
How tightly the agent is bounded and how you know it stays good. The base carries baseline guardrails; these harden and measure them.
How a person stays in control. The base pauses and alerts you when unsure; these add routing, a review console, and approval gates.
What you can see once the agent is live. The base logs every run; these turn logs into tracing, dashboards, and cost control.
Where the agent runs and how it holds up under load. Managed cloud is the fast path; self-hosting or a private model keeps everything on your infrastructure.
What happens after launch: getting your team running it, extending the warranty, a service commitment, and anything not on this sheet.
How delivery runs
- 1Access and knowledgeThe systems, credentials, and the starter materials the agent is grounded in. A few days, mostly on your side.
- 2Build and evaluationBuilt and measured against your real scenarios on a staging setup, before it ever touches production.
- 3AcceptanceYou confirm it on your own cases against the agreed scope.
- 4Go-live and warrantyIt runs on your own domain, documented and yours, with a 30-day warranty.
Third-party running costs
- Model usage (API)$10-200/moBy conversation volume and the model chosen; a self-hosted or private model shifts this to your own infrastructure bill.
- Retrieval store$0-50/moFor retrieval over your documents (RAG); a free tier covers a small knowledge base.
- Hosting$5-40/moManaged cloud on your own domain; a small always-on service in an account you own.
- Connected toolsvariesAny paid plans your own systems already require; billed by those providers directly.
Paid by you directly to the providers, never through SporeSec. Your accounts, your control, no dependence on us to keep running.
Need a full custom system around the agent, not just the agent itself? A custom build starts from ≈$8,330. See Custom Build
Review your build
Remove anything you don't need before you save.
- Autonomous AI agent (base)≈$8,330
- Knowledge base: Retrieval over your documents (RAG)+≈$1,500
- Tool integration+≈$840
- Memory model: Long-term memory+≈$1,000
- Guardrail tier: Strict+≈$1,000
- Conversation tracing+≈$1,000
- Analytics dashboard+≈$840
- Cost monitoring and budgets+≈$500
We build it. We look after it.
Free, 30 minutes, no sales deck. We ask about the problem, the goal, and what has been tried. If we fit, you get a fixed-price proposal.
- Free intro call30 minutes, free. No pitch.
- AI and automation call30 minutes, free. AI agents, WhatsApp, CRM and workflows.
- Website and online store call30 minutes, free. WordPress, WooCommerce, Shopify or a custom build.
- IT and security call30 minutes, free. Computers, accounts, email, backups and security.