Frequently Asked Questions

General

What is Supervity?

Supervity is a self-driving operations platform where AI agents act as autonomous employees — planning, executing, and monitoring work under human-defined policies and approvals.

You describe the outcome you want, review how the agent plans to achieve it, and approve execution.
Automation happens without losing human control.

Is Supervity just another automation or workflow tool?

No.

Traditional automation tools require you to manually design every step.
Supervity uses AI agents that reason, plan, and adapt workflows automatically, while humans remain in command through approvals, constraints, and oversight.

Think of it as delegating work to an AI employee, not wiring rules and triggers.

Will Supervity replace my team?

No.

Supervity is designed to augment teams, not replace them.

Agents handle repetitive, operational, and error-prone work so humans can focus on:

• judgment and decision-making
• strategy and planning
• creativity and exceptions

Humans remain accountable for outcomes.

Do I need to know how to code?

No.

Most users interact with Supervity using natural language.
Developers can optionally integrate via APIs.

You can start without any technical background.

What kinds of problems is Supervity best suited for?

Supervity works best for:

• repetitive operational processes
• cross-tool workflows (email, CRM, files, data)
• approval-based or compliance-sensitive tasks
• processes that need monitoring and recovery
• tasks that benefit from AI reasoning, not rigid rules

Getting Started

How do I get started?

1. Sign up and create a workspace
2. Connect the tools your agents will use
3. Describe what you want an agent to do
4. Review and approve the plan
5. Let the agent run

See the Quick Start Guide for a walkthrough.

How long does it take to see value?

• simple use cases: 5–10 minutes
• multi-step operational flows: 30–60 minutes

Most teams see value on the first day.

Do I need to install anything?

No.
Supervity is fully cloud-based and runs in your browser.

Working with Agents

How much control do I have over AI agents?

Full control.

Agents:

• cannot act without approval (unless explicitly allowed)
• operate only within permissions you grant
• pause when human input is required
• produce complete logs and audit trails

This is a human-in-command model, not blind automation.

What happens if an agent makes a mistake?

Agents are designed to fail safely.

They:

• retry where appropriate
• surface errors clearly
• pause for human intervention when needed

You can review logs, correct inputs, update constraints, and re-run safely.

Nothing is irreversible by default.

Can agents run continuously without supervision?

Yes — once approved.

Agents can:

• run on schedules
• respond to events
• be triggered via APIs

You still retain:

• alerts
• logs
• approval gates
• the ability to pause or stop execution

Human Review & Approvals

When should I use human review?

Use human review for:

• financial approvals
• external communications
• compliance-sensitive actions
• high-risk operations
• edge cases and exceptions

Supervity lets AI do the work until judgment is required.

Who can approve or review actions?

You define reviewers.

They can be:

• specific users
• roles (manager, admin)
• dynamically assigned based on data

Learn more in Human Review & Approvals.

Can reviews happen outside Supervity?

Yes.

Review forms can be:

• accessed via email
• embedded in dashboards
• integrated into external systems using APIs

Integrations & Data

What integrations are available?

Supervity supports 50+ integrations, including email, chat, storage, CRM, and databases.

Agents use integrations as tools to perform work.

See Integrations Overview.

What data does Supervity access?

Only what you explicitly authorize.

Agents:

• do not have blanket access
• operate with least-privilege permissions
• cannot access systems you haven’t connected

Access can be revoked at any time.

Execution & Reliability

What happens if something fails mid-execution?

Supervity supports:

• automatic retries
• timeouts
• fallback paths
• notifications
• manual intervention

Failures are visible and recoverable, not silent.

How long can workflows run?

• default timeout: 30 minutes
• extended limits available on higher plans

Long-running workflows are supported with checkpoints.

Security & Compliance

Is Supervity safe for enterprise use?

Yes.

Supervity is designed for enterprise environments with:

• encryption in transit and at rest
• role-based access control
• audit logs
• compliance-aligned architecture (SOC 2, GDPR, HIPAA, etc.)

See Security & Privacy.

Can Supervity employees see my data?

No — unless you explicitly grant access for support.

All access is logged and auditable.

APIs & Developers

Does Supervity have APIs?

Yes.

You can:

• trigger agents programmatically
• monitor execution
• embed human-review flows
• integrate with existing systems

See API Reference.

Can I embed Supervity into my own application?

Yes (Enterprise).

Supervity supports:

• API-driven execution
• embedded approval forms
• external command centers

Pricing & Plans

What counts as an execution?

One complete agent run counts as one execution — regardless of how many steps it contains.

Can I change plans later?

Yes.
You can upgrade, downgrade, or cancel at any time.

Troubleshooting & Support

My workflow isn’t behaving as expected. What should I do?

1. check execution logs
2. review agent plan and inputs
3. verify integrations and permissions
4. test manually
5. contact support if needed

See Troubleshooting.

Where can I get help?

• documentation and tutorials
• community forum
• email support
• enterprise plans include dedicated support and SLAs

Still Have Questions?

If you can’t find what you’re looking for:

• search the docs
• ask in the community
• email [email protected]

We’re here to help.