General Assistantby Lindy★ 4.3

Lindy

Build AI assistants for ops and customer-facing workflows. Lindy's strong on email, scheduling, and CRM-style automations.

Try it

Safety & Trust

Delv Safety Grade: C

Score 58/100 · assessed 2026-04-18

Maintainer65

Permissions40

Supply chain50

Transparency35

Incidents100

Lindy is a commercial AI automation platform from a funded startup (Y Combinator W23). It's positioned between workflow tools and autonomous agents, handling email triage, scheduling, and CRM operations. The maintainer is a legitimate venture-backed company with reasonable track record, but transparency is poor: no public repository, closed-source architecture, and minimal technical documentation about how it processes sensitive business data. Permissions are broad by necessity (email read/write, calendar access, CRM integration, external LLM calls), creating significant attack surface. Supply chain is opaque since everything runs on Lindy's infrastructure with no package to audit. The freemium model means you're trusting a third party with customer emails and calendar data before they've proven long-term viability. No known incidents, but the closed nature and broad access make this a 'trust the vendor' proposition rather than a verifiable one.

Green flags

Y Combinator backed with identifiable founding team
Focused use case (email/scheduling) rather than claiming to do everything
No known security incidents or breaches to date
Active product with regular updates per user reports

Red flags

Closed source with no repository or technical architecture docs
Broad email and calendar access with opaque data handling
Freemium SaaS model creates unclear long-term data custody
No audit trail or security certifications mentioned publicly
External LLM processing of potentially sensitive business communications

Permissions requested

Read messagesSend messagesIdentity readIdentity writeOutbound networkExternal LLM callAccess secrets

Assessed by Delv Editorial using public metadata. Grades are advisory and update as the ecosystem changes. They do not replace your own review of permissions and code before granting an agent access to sensitive systems.

Pricing

FREEMIUM

Platforms

web

Review

Lindy sits in the awkward middle ground between Zapier and a full-blown AI agent. It's not quite autonomous enough to ignore for days, but it handles repetitive email and scheduling work better than anything that requires you to click "approve" every step. I've run it on inbound lead triage for a SaaS product. You feed it rules ("if email mentions enterprise, tag urgent and book a call"), connect your calendar and CRM, and it actually does the boring bits. The scheduling piece is genuinely useful: it reads back-and-forth threads, proposes times, sends calendar invites, and doesn't need babysitting once you've trained it on your availability preferences. That's where the autonomy pays off. A tool like ChatGPT would make you paste each email, review the draft, copy the response back. Lindy just does it. The CRM integrations are competent but not magical. It'll log interactions in HubSpot or Salesforce, update lead status, fire off Slack pings. You're not getting deep data enrichment or predictive scoring. It's more "keep the CRM current without manual data entry" than "replace your ops team." Failure modes: it occasionally misreads tone in customer emails, especially sarcasm or urgency buried in polite language. I caught it booking a demo with someone who was actually asking to cancel. The rule builder is visual and fairly intuitive, but complex conditional logic gets messy fast. If your workflow has more than four or five branches, you'll spend time debugging why it took the wrong path. Compared to something like Relay.app, Lindy's AI layer is stronger but the integration catalogue is narrower. Relay gives you more connectors and better version control on workflows. Lindy gives you an agent that reads context and adapts without you rewriting the whole automation when edge cases appear. The freemium tier is generous enough to test a real workflow. Paid plans start to sting if you're running multiple assistants, but the per-seat pricing makes sense for small teams where one person manages several bots. This isn't the agent that's going to run your entire ops stack unsupervised. It's the one that saves you two hours a day on email triage, meeting scheduling, and CRM hygiene. If that's your pain point, Lindy's worth the setup time. If you need deep reasoning or multi-step research, look elsewhere.

Verdict

Pay for Lindy if you're drowning in scheduling emails and CRM busywork and you trust an AI to handle low-stakes customer interactions. Skip it if you need complex reasoning, broad integrations, or genuinely autonomous decision-making beyond rule-based triage.

Good at

Scheduling autonomy actually works - reads threads, proposes times, sends invites without supervision
Email triage and response quality is a step above basic automation tools
Generous freemium tier lets you test real workflows before paying
CRM logging happens reliably once configured, saves manual data entry
Rule builder is visual and doesn't require coding for straightforward workflows

Watch out

Occasionally misreads tone or urgency in customer emails, leading to wrong actions
Integration catalogue narrower than Zapier or Relay.app
Complex conditional logic gets messy and hard to debug
Paid tiers get expensive if you're running multiple assistants
Not autonomous enough for true set-and-forget - still needs periodic supervision

Use cases

Inbound lead triage
Auto-scheduling sales calls
Customer support deflection
Internal ops bots