AI Dispatch
The wirewireAug 28, 2026

Workflow agent Lindy requires a named approver before irreversible actions; discloses no insurance or audit-log retention terms

The vendor's site describes Slack-channel scoping and a required human sign-off unusually specific for the category, then publishes no liability cap, indemnification term or audit-trail retention policy.

Lindy, an agent that says it operates in "Slack, iMessage, or in your browser," publishes governance detail unusual for a workflow-automation product, according to its own site. "Anything with outside impact waits for a named approver: sending an email, updating a ticket, posting to another channel, publishing a doc," the site states, adding: "Nothing irreversible happens without your approval."

The same specificity extends to data access. On Slack permissions, Lindy's site says the agent sees "Public channels once it's in your workspace. Private channels only if a channel manager invites it. It never sees archived channels, and never anyone's personal DMs." The company connects to "Gmail, Slack, Notion, HubSpot, and 1,000+ more. Supports MCP," and says "Your data is never sold and never used for training."

On compliance, Lindy self-declares "SOC 2 Type II, GDPR, HIPAA, and PIPEDA, encrypted in transit and at rest," and states "Enterprise adds HIPAA with a signed BAA." As checked, the site names no certificate number, auditor or report date behind any of the four frameworks. Its audit-trail disclosure is a single feature-list entry reading "Audit logs," with no retention period, scope or tamper-evidence claim attached.

Pricing tracks usage rather than results. Lindy's site says "Plans start at $29.99 per user per month with 3,000 credits" pooled across a workspace, a structure that charges the same amount whether or not the work performed is used. No AI liability insurance policy, named carrier, policy limit, customer indemnification term or outcome-linked payment mechanism is published on the pages checked.

Lindy's approval-gate and channel-scoping disclosures put it ahead of most agent products aidispatch.news has audited on runtime governance — the dimension of the trust-gap rubric that measures whether scope, spending or authority limits are enforced while an agent works, rather than asserted in marketing copy alone. The absence of insurance and audit-retention terms, by contrast, is the same gap found across the category.

Entries in this piece 1

Published index entries backed by the same source documents this piece cites.

Sources 10

  1. Whether it's Slack, iMessage, or in your browser, Lindy is ready to pitch in.
    lindy.ai · checked Aug 10, 2026
  2. Anything with outside impact waits for a named approver: sending an email, updating a ticket, posting to another channel, publishing a doc.
    lindy.ai · checked Aug 10, 2026
  3. Nothing irreversible happens without your approval.
    lindy.ai · checked Aug 10, 2026
  4. Public channels once it's in your workspace. Private channels only if a channel manager invites it. It never sees archived channels, and never anyone's personal DMs.
    lindy.ai · checked Aug 10, 2026
  5. Gmail, Slack, Notion, HubSpot, and 1,000+ more. Supports MCP, so the world is your oyster.
    lindy.ai · checked Aug 10, 2026
  6. Your data is never sold and never used for training.
    lindy.ai · checked Aug 10, 2026
  7. SOC 2 Type II, GDPR, HIPAA, and PIPEDA, encrypted in transit and at rest.
    lindy.ai · checked Aug 10, 2026
  8. Enterprise adds HIPAA with a signed BAA.
    lindy.ai · checked Aug 10, 2026
  9. Audit logs
    lindy.ai · checked Aug 1, 2026
  10. Plans start at $29.99 per user per month with 3,000 credits.
    lindy.ai · checked Aug 10, 2026