Create your own AI helper in minutes.
Describe what you need in plain words, connect the apps you already use, and let your AI agent do the work. It always asks you before doing anything important, and shows you exactly what it did.
Free sign-up, no card needed. Get 1,000 free AI credits.
An easy way to put AI to work, safely
Voidex lets anyone create AI agents that work with the apps they already use. You describe the job in your own words, connect your apps, and decide what the agent is allowed to do on its own. For anything important, like sending a message or changing a record, it asks you first.
After every task, Voidex checks that it really happened and keeps a clear record you can look back on. No coding needed. Developers can also use Voidex through its API.
Three simple steps
Describe your agent
Give it a name and tell it what to help with, in plain words. For example: “Answer customer questions and book appointments.”
Connect your apps
Pick the apps you use, like email, spreadsheets, your shop or your CRM. Any app with an MCP connection works.
Stay in control
Your agent asks before doing anything important. You see everything it did, and you can pause it any time.
What happens every time your agent does something
Your rules are checked first. If you asked to approve it, the agent waits for your OK. Then the task runs, and Voidex checks the result in the real app before calling it done. Scroll to follow one task.
- 1. Action
An agent, workflow or API call proposes a typed action. Input is validated against the tool’s JSON Schema before anything else happens.
- 2. Policy
- 3. Approval
- 4. Execute
- 5. Verify
- 6. Receipt
POST /api/v1/runs
{ "tool": "acme_ledger.create_invoice",
"connectedAccountId": "acct_…",
"input": { "customer": "Globex", … } }A clear record of everything your agent does
- Checked, not assumed. After your agent does something, Voidex looks in the real app to make sure it actually happened.
- Honest results. If something went wrong, you see it clearly. A problem is never shown as “done”.
- No double actions. If an app doesn’t answer, Voidex checks first instead of trying again, so nothing is sent or charged twice.
- Records you can trust. Every record is sealed so it can’t be quietly changed later.
Horizontal by design, honest about status
Generic REST / HTTPS
AvailableAny JSON API you control, behind host allowlists and SSRF protection.
Signed outbound webhooks
AvailableHMAC-SHA256 with timestamps; retries with backoff.
MCP endpoint
AvailableYour registered tools exposed to MCP clients — still policy-checked.
Acme Test CRM & Ledger
SimulatedLocal simulators with fault injection for development and Bench.
SaaS connectors (CRM, support, dev tools, ERP)
PlannedSalesforce, HubSpot, Zendesk, GitHub, Jira, ERPNext, Odoo and others are on the roadmap.
Model providers
PlannedProvider adapters are interface-ready; only a deterministic demo provider ships today.
Safe by default
Nothing happens without permission
Your agent only does what you allow. Big or risky actions need approval, and the most serious ones need two people.
Your data stays yours
Each account is kept separate, and team members only see what their role allows.
Passwords stay locked
App passwords and keys are encrypted and never shown again, not even to the AI.
No surprises
Your agent can only use the app actions you connected. It cannot run unknown programs.
Common questions
- What is Voidex?
- Voidex lets anyone create AI agents that work with the apps they already use, without coding. Under the hood it is infrastructure for AI agents that take actions in business systems. It connects agents to typed tools, enforces deny-by-default policy and human approvals before anything runs, executes actions in an isolated worker, then independently checks the real system to confirm what actually happened and issues a tamper-evident receipt.
- How does Voidex verify an AI agent’s action?
- After an action executes, Voidex runs typed postcondition checks that read the target system independently, for example confirming a record exists, a field equals the requested value, line items sum to a total, or a status moved from draft to posted. Each check is passed, failed, inconclusive or not run, and only passed deterministic or human checks count as verified.
- How do human approvals work?
- Policies can require approval for sensitive actions. An approval is bound to a hash of the exact payload, target and policy version, and it expires. If anything changes after approval, the approval no longer applies. Requesters cannot approve their own actions, and high-impact actions always need two different approvers.
- Does Voidex support MCP (Model Context Protocol)?
- Yes. Voidex exposes a workspace’s registered tools through an MCP server endpoint over streamable HTTP, authenticated with an API key. MCP is only a transport: every MCP tool call goes through the same policy, approval, verification and receipt path as the REST API. Connecting Voidex to external MCP servers as a client is planned.
- Is Voidex SOC 2 or ISO 27001 certified?
- No. Voidex holds no compliance certification. It provides controls such as role-based access, approvals, append-only audit logs, receipts and retention settings that can support a customer’s own compliance work, and it publishes which controls are implemented and which are planned.
Create your first AI agent today
Free sign-up, no card needed, 1,000 free AI credits. Or tell us what you need and we will build it for you.