Build AI workflows and agents in natural language — 200+ app integrations, multi-model AI steps, and human approval built into every run.
Task: Automating WorkflowsUse case: Build and automate ai powered workflows across 200 apps using natural languageModels: claude (opus 4.8), custom models via api key, bedrock
Accounts for apps to connect (Gmail, Google Drive)
Airtable base (for CRM-style automations)
Calendly account (for scheduling triggers)
Slack workspace (for approval notifications)
Skill level
No code — point and click
Worth knowing: Fewer native app integrations than Zapier/Make, though it covers the high-volume tools most people already use.
Summarized from 2 community tutorials · July 2026
Highlights
Natural-language AI workflow builderDescribe a workflow in plain English to the relay.app assistant chatbot; it structures a trigger + AI + action plan and asks clarifying questions before buildin
Human-in-the-loop Slack approvalsPlaybooks can pause mid-run to send a Slack notification with an AI-drafted email so a person can review, edit, or approve before it sends.
Built-in AI text steps (summarize/extract/generate)Individual playbook steps can call AI to summarize, extract, or generate text from data pulled out of a connected app, with adjustable length and hint context.
Conditional branching on record lookupsPlaybooks branch into different paths (e.g., contact exists vs. does not exist in Airtable) and map fields between connected apps without writing code.
No-API-key native integrationsConnects directly to common tools (Calendly, Airtable, Slack, Gmail, Google Drive, HubSpot, Notion) via authenticated app connections rather than requiring API
Summarized from 2 community tutorials · July 2026
i
No approved reviews yet. Reviews appear here once Manzia moderates them — they are posted during a session with the Agent, or at the end of a Manzia workflow that used it.
