Lindy review: an AI assistant across inbox, calendar, Slack and phone

Contents

1. Approvals and their reach
2. Security and data
3. How it is billed
4. Strengths and limits
5. Pricing
6. Questions to ask Lindy
7. Sources

Type	AI assistant platform
Pricing	Published
Website	lindy.ai

Lindy presents itself as an AI employee that plugs into a company's tools and takes work off people's plates. It connects to more than 1,500 integrations and any hosted MCP server, works across email, calendar, Slack, text and phone, takes meeting notes, and can drive a browser or write its own integration where none exists⁠12Lindy: hire an AI employee that gets work doneLindyIntegrationsLindy Docs.

Best for: Teams that want an AI assistant working across email, calendar, Slack, text and phone in the tools they already use, with approval before anything leaves the company.

Not the right fit for: Companies that need approvals enforced in every channel, self-hosting or a regional data option.

Approvals and their reach

Lindy's default is that anything with outside impact waits for approval, while read-only lookups do not⁠3PricingLindy. Guardrails are set per integration with three options: always allow, require approval (Lindy asks in Slack before writing) or don't offer. Approve and Deny buttons appear in the thread; the requester or a workspace owner or admin can approve, and approval covers only that one action⁠2IntegrationsLindy Docs.

The documentation is candid about the limits. Guardrails apply in shared Slack threads only, not in the web app, Slack direct messages or text messages. A guardrail covers every account under an integration, and connecting a tool turns on all of its actions at once⁠2IntegrationsLindy Docs. For a company that wants every write approved regardless of channel, that is a material gap.

Security and data

Lindy states that it is SOC 2 Type II, GDPR, HIPAA and PIPEDA compliant, encrypts data in transit and at rest, and does not sell data or use it to train models. SSO, SCIM, audit logs and HIPAA with a signed business associate agreement are Enterprise features⁠43SecurityLindy DocsPricingLindy. Self-hosting and regional data options are not stated.

How it is billed

Every active user is billed from $29.99 a month for 3,000 credits, drawing on one shared pool. Lindy publishes typical credit costs: everyday requests use 2 to 250 credits, deeper work 250 to 1,000 and large builds 1,000 to 2,500. Monthly credits do not roll over, and when the pool runs out, credit-using actions pause until the next cycle⁠3PricingLindy.

Strengths and limits

Strengths

• Safe default: actions with outside impact wait for approval; read-only lookups do not.
• Guardrails per integration with three settings (always allow, require approval, don’t offer), approved with buttons in the Slack thread.
• Covers the widest set of channels for a personal or team assistant, including phone calls and meeting notes.
• Credit costs per kind of task are published, and work pauses at the limit rather than producing a surprise bill.

Limits

• Guardrails apply only in shared Slack threads, not in the web app, Slack direct messages or text messages.
• Connecting a tool enables all of its actions at once, and one guardrail covers every account under an integration.
• SSO, SCIM and audit logs are Enterprise features; self-hosting and data residency are not stated.

Choose another option if: Approval must hold in every channel, or agents must work on records in a system you control. OpenLazarus or Relevance AI fit better.

Pricing

Published: a free week with $50 in credits; Team from $29.99 a month per active user for 3,000 credits; top-ups $10 per 1,000 credits.

Questions to ask Lindy

• Which of our channels would bypass the approval guardrails, and can we disable them?
• Can we enable only some actions of an integration, such as reading but not sending?
• Which plan gives us audit logs, and how long are they kept?
• Where is our data stored?

See how Lindy compares with the other agent builders in the main ranking.

Sources

1. Lindy: hire an AI employee that gets work done. Lindy. Accessed 8 Nov 2026.
2. Integrations. Lindy Docs. Accessed 8 Nov 2026.
3. Pricing. Lindy. Accessed 8 Nov 2026.
4. Security. Lindy Docs. Accessed 8 Nov 2026.

Further reading

• Buyer's guide · Updated quarterlyThe best no-code AI agent builders for company work in 2026An analysis of seven platforms for building AI agents without code, judged on the controls a company needs before an agent changes a real order, invoice or customer record.
• Guide · ControlAI agent permissions and approvals: how to let agents act safelyAn agent's permissions are a job description written in advance. The platforms differ less in whether they offer approvals than in where those approvals apply and what record they leave.
• Guide · CostWhat no-code AI agents cost: credits, actions, executions and the monthly billEvery platform meters agent work in its own unit. Converting those units into the cost of one task is the only way to compare them, and the only way to avoid the cost overruns Gartner expects to end many projects.