Question 1

What is Agento?

Accepted Answer

Agento is an enterprise AI workflow automation platform, built in Melbourne, Australia. You describe a business process in plain language, and Agento's governed AI agents run it end to end across the tools you already use, with a human approving anything sensitive and every action recorded to a tamper-evident audit trail. Teams use it in construction, financial services, HR and compliance.

Question 2

Who is Agento for?

Accepted Answer

Agento is built for teams in regulated and evidence-heavy industries: construction, financial services, healthcare, energy, and any business with compliance obligations. It suits organisations that want AI agents doing real operational work but can't accept business process automation they're unable to control, explain or audit afterwards.

Question 3

How is Agento different from Zapier, Make or n8n?

Accepted Answer

Zapier, Make and n8n connect apps using fixed rules: if this happens, do that. Agento runs agentic AI workflows that handle judgement and exceptions, not just triggers, and wraps them in the governance a regulated business needs: scoped permissions, human-in-the-loop approval on sensitive steps, durable execution that resumes after a failure, and a tamper-evident record of every action.

Question 4

How is Agento different from RPA tools like UiPath or Automation Anywhere?

Accepted Answer

RPA bots repeat scripted clicks and break when a screen changes. Agento's AI agents work from the intent of a process rather than a fixed script, so they handle variation and exceptions without re-recording. Where you do need direct application control, an Agento Operator drives the software while every action stays governed and logged.

Question 5

What is an Agento skill?

Accepted Answer

A skill is a business process packaged once and reused everywhere: invoice extraction, subcontractor onboarding, RFI routing. You build it in a no-code visual editor, test it against sample data, and promote it from draft to staging to production with full version control. Teams share skills rather than rebuilding the same automation repeatedly.

Question 6

What is an Agento Operator, and can it automate software with no API?

Accepted Answer

An Operator is a computer-use AI agent that drives an application directly, clicking, typing and reading the screen, for software with no integration available. It works desktop applications like AutoCAD, legacy systems and web portals, following your process step by step. Every click and keystroke is recorded, so the work stays as auditable as any API-based step. Browser operators are available from the SME tier; desktop and virtual-desktop operators run in isolated virtual machines on Enterprise.

Question 7

Do I need engineers or developers to use Agento?

Accepted Answer

No. Agento is no-code AI workflow automation: you describe what you want in plain language and build it in a drag-and-drop editor. Most teams get their first workflow running without engineering involvement. Developers can extend Agento through the API when you need something bespoke, but that's optional rather than a prerequisite.

Question 8

How does Agento work?

Accepted Answer

You describe the workflow in plain language. Agento builds it as a multi-step AI workflow with branching logic, human approval steps and error handling, then runs it across your connected tools. If a step is interrupted, the workflow resumes exactly where it stopped and finishes on its own. Nothing to restart, nothing lost.

Question 9

What apps and integrations does Agento connect to?

Accepted Answer

Agento connects to 50+ enterprise integrations, including Salesforce, Microsoft 365, Google Workspace, Slack, Teams, SharePoint, SAP, HubSpot, Jira, AWS and GitHub, plus Procore, Autodesk, AutoCAD, InEight and HammerTech for construction teams. You can connect additional apps through our API, and where a tool has no integration at all, an Agento Operator can drive it directly. Enterprise tenants support service-account and mTLS connection patterns, with a documented security review per connector.

Question 10

What if the tool I use isn't supported?

Accepted Answer

Two options. You can connect it yourself through our API or the connector builder, or we can add it: we build new integrations quickly and prioritise by customer demand. If the software has no API of any kind, an Agento Operator can work the application directly instead, so a missing integration never blocks automation.

Question 11

How are workflows triggered?

Accepted Answer

On a schedule, on an event in a connected system, or on demand when someone runs them. An AI workflow can wait on a human approval, branch on a condition, run steps in parallel, and hand off between systems, all within a single defined process.

Question 12

How long does it take to get a first workflow live?

Accepted Answer

It depends on the tier and the governance you need. An individual describing their own process is live the same day. A team on the SME tier is typically running in one to two weeks, including assisted setup and a shadow-mode run. An Enterprise deployment takes about five weeks of delivery (discovery, connectors, policy encoding, shadow mode and change control) or six to twelve weeks end to end once your own security review is included. Later workflows are much faster, because connectors and policy are already in place.

Question 13

Which AI models does Agento use?

Accepted Answer

Agento routes each step to the AI model best suited to it, balancing reasoning, speed and cost, so you don't have to manage model choice yourself. You're not locked to a single provider. Enterprise tenants get reserved routing capacity and can constrain routing by data residency, provider or certification, or connect their own private models.

Question 14

Who approves what the AI does?

Accepted Answer

You do. Agento is human-in-the-loop by design: you define which steps require sign-off (payments over a threshold, anything customer-facing, whatever your policy requires) and those steps pause until a named person approves them. From the SME tier, an approver cannot be the person who initiated the action. Enterprise encodes your full delegation of authority as an approval matrix by action, risk tier and role.

Question 15

Can I restrict which systems and data an AI agent can access?

Accepted Answer

Yes. Every connector carries an explicit permission scope, so an agent reaches what the workflow needs and nothing beyond it. Credentials are held in a secrets store and injected at execution; they are never written into a workflow definition and never exposed to a model. Enterprise tenants also define their own network egress allowlist per profile.

Question 16

What happens if an AI agent makes a mistake?

Accepted Answer

Sensitive actions don't execute without approval, so most mistakes are prevented rather than corrected. Where something does go wrong, the audit trail shows exactly what ran, what it touched and what was approved. AI guardrails score risk in real time and escalate automatically when a threshold is crossed, and Enterprise workflows carry a formal change record against the workflow version hash with a rollback window.

Question 17

What happens if a workflow fails halfway through?

Accepted Answer

It resumes from where it stopped. Agento's execution is durable, so an interrupted workflow picks up at the exact step it left off and completes on its own. No partial work lost, nothing to restart manually, and no one needing to monitor it.

Question 18

Can I test a workflow before it runs on real data?

Accepted Answer

Yes. Every skill can be tested against sample data, and workflows move through a draft, staging and production pipeline with version control. Before go-live, workflows run in shadow mode (5 to 20 runs on the SME tier, 50 to 500 on Enterprise) against an agreed concordance threshold, so you can compare what the agent would have done against what your team actually did.

Question 19

How is security and AI governance handled?

Accepted Answer

Every tier, including the trial, carries the same security floor: authenticated identity on every action, scoped connector permissions, credentials brokered from a secrets store and never exposed to a model, immutable execution records with an integrity hash, TLS 1.3 in transit and AES-256 at rest with a separate encryption key per tenant, sandboxed execution, and row-level security enforced in the database so a query cannot cross a tenant boundary. Governance is configurable. The floor is not.

Question 20

Where is my data stored?

Accepted Answer

Individual accounts are hosted in Australia. SME tenants are provisioned in a single region chosen at setup. Enterprise tenants choose their region, with a per-region tenant boundary, and can use customer-managed encryption keys. Data is encrypted with TLS 1.3 in transit and AES-256 at rest, using a distinct data-encryption key for each tenant.

Question 21

Is my data used to train AI models?

Accepted Answer

No. Your data is used to run your workflows and is not used to train models, ours or any provider's. Credentials are never exposed to a model at all.

Question 22

What's the difference between shared and dedicated tenancy?

Accepted Answer

Individual and SME tenants share infrastructure and are separated logically, by tenant identifier, row-level security, table partitioning and a per-tenant encryption key. Enterprise tenants are separated physically: a dedicated database instance, an isolated execution pool, a dedicated storage bucket, separate backups and their own subdomain at yourcompany.app.agento.au. That address is the visible edge of a real boundary, not branding.

Question 23

How does the audit trail and evidence capture work?

Accepted Answer

Every AI action emits a structured record with an integrity hash: screenshots, activity logs, transcripts, and the files produced along the way. Records can be exported and deleted on request, but cannot be silently edited. Enterprise includes an auditor-ready evidence pack per run or per period, and a write-once retention option.

Question 24

How long is evidence retained, and can we export it?

Accepted Answer

Individual retains evidence for 7 days. SME retains 12 months, configurable. Enterprise is configurable to 7 years or more with a write-once option. You can export a complete audit report at any time, with the full evidence chain attached, in a form you can hand directly to an auditor or regulator.

Question 25

Does Agento support SSO, SAML and SCIM?

Accepted Answer

Sign-in uses OIDC with Google or Microsoft accounts on every tier, with multi-factor inherited from your identity provider. Enterprise adds SAML against your own IdP, mandatory SCIM provisioning so joiners and leavers are handled automatically, conditional access, custom roles, and just-in-time privilege elevation that is time-boxed and recorded.

Question 26

Can we self-host or run Agento in our own cloud?

Accepted Answer

Yes, on Enterprise. Agento can be deployed into your own cloud account rather than hosted by us. Enterprise tenants can also constrain model routing by region, provider or certification, and define their own network egress allowlist, so you control both where data sits and where it can go.

Question 27

Can Agento handle regulated data?

Accepted Answer

Regulated data isn't permitted on the Individual tier. SME permits it with restrictions agreed at onboarding. Enterprise supports it with data classification and controls evidenced: a negotiated DPA, a BAA where applicable, per-connector and per-workflow data-flow documentation reviewed with you, and the control bundles you require.

Question 28

What is your SOC 2 and ISO certification status?

Accepted Answer

Agento applies SOC 2-aligned controls and an ISO/IEC 42001-aligned management approach. Formal SOC 2 attestation and ISO/IEC 42001 certification are in progress and not yet held. Penetration test summaries are available under NDA, and on an annual cadence for Enterprise. Enterprise customers get full security questionnaire support, including an architecture walkthrough. Contact security@agento.com.au.

Question 29

Do you have a DPA and a sub-processor register?

Accepted Answer

Yes. Standard terms apply on Individual, a standard DPA on SME, and a negotiated DPA with a BAA where applicable on Enterprise. Our sub-processor register is published, with change notification on SME and an objection window on Enterprise.

Question 30

What happens to our data if we stop using Agento?

Accepted Answer

Individual and SME accounts support immediate self-service deletion. Enterprise deletion is policy-driven with a legal-hold override. You can export your workflows, evidence records and audit history at any time before deletion, and we'll confirm deletion in writing on request.

Question 31

How much does Agento cost?

Accepted Answer

Agento has three tiers. Individual is free for 7 days, then A$20 per user per month, self-serve, with no sales conversation. SME and Enterprise are quoted, because the honest number depends on how many systems you need to reach and how much evidence you need to keep. Tell us those two things and we'll give you a number in the first conversation.

Question 32

What's included in the 7-day trial?

Accepted Answer

Full capability at the Personal assurance lane, not a limited demo. You can connect your own accounts, describe a workflow in plain language, run it, and see the complete evidence trail. Browser and desktop operators aren't included in the trial, and regulated data isn't permitted on the Individual tier.

Question 33

Do I need a credit card to start?

Accepted Answer

No. No payment details are collected at signup, so nothing can be charged automatically. On day seven the workspace asks you to subscribe: workflows pause, and nothing is deleted for 30 days. You'll get reminders on day five and day seven.

Question 34

Can I cancel at any time?

Accepted Answer

Yes, in one click inside the product, at any time. No phone call and no retention flow.

Question 35

Why isn't SME and Enterprise pricing published?

Accepted Answer

Because a single published number would be wrong for most enquirers in both directions. SME cost varies with connector count, run volume and whether operators are needed. Enterprise is provisioned infrastructure (a dedicated database, an isolated execution pool, reserved model-routing capacity) so it has a real floor that varies with the deployment. We'll scope it and give you a number in the first conversation.

Question 36

What do I get by moving from SME to Enterprise?

Accepted Answer

Physical rather than logical isolation: your own database instance, execution pool, storage bucket and subdomain. Plus SAML and mandatory SCIM, customer-managed encryption keys, region choice, desktop and virtual-desktop operators, policy-as-code enforcing your delegation of authority, retention to 7 years or more, and auditor-ready evidence packs. Nothing is thrown away on the way up: workflows, skills and evidence migrate, because the assurance lane changes rather than the platform.

Questions, answered.

About Agento

How it works

Control and reliability

Security, data and compliance

Plans and pricing

Still have questions?

Questions, answered.

About Agento

1What is Agento?

2Who is Agento for?

3How is Agento different from Zapier, Make or n8n?

4How is Agento different from RPA tools like UiPath or Automation Anywhere?

5What is an Agento skill?

6What is an Agento Operator, and can it automate software with no API?

7Do I need engineers or developers to use Agento?

How it works

1How does Agento work?

2What apps and integrations does Agento connect to?

3What if the tool I use isn't supported?

4How are workflows triggered?

5How long does it take to get a first workflow live?

6Which AI models does Agento use?

Control and reliability

1Who approves what the AI does?

2Can I restrict which systems and data an AI agent can access?

3What happens if an AI agent makes a mistake?

4What happens if a workflow fails halfway through?

5Can I test a workflow before it runs on real data?

Security, data and compliance

1How is security and AI governance handled?

2Where is my data stored?

3Is my data used to train AI models?

4What's the difference between shared and dedicated tenancy?

5How does the audit trail and evidence capture work?

6How long is evidence retained, and can we export it?

7Does Agento support SSO, SAML and SCIM?

8Can we self-host or run Agento in our own cloud?

9Can Agento handle regulated data?

10What is your SOC 2 and ISO certification status?

11Do you have a DPA and a sub-processor register?

12What happens to our data if we stop using Agento?

Plans and pricing

1How much does Agento cost?

2What's included in the 7-day trial?

3Do I need a credit card to start?

4Can I cancel at any time?

5Why isn't SME and Enterprise pricing published?

6What do I get by moving from SME to Enterprise?

Still have questions?