Folder
Guides
Guides combine Builders triggers, processing blocks, integrations, data resources, and runtime controls into complete automations. Each guide starts with an outcome, builds the workflow in execution order, and finishes with a live verification and an operational checklist.
Available guides
- Build an Automated Email Response Workflow - receive a new message, read its full content, generate a structured draft, and reply through the same mail setup.
- Process Uploaded Files from an Endpoint - accept one multipart file, validate and transform its content, persist the result, and return a synchronous API response.
- Advanced File Processing - stage mixed uploads, classify bounded batches on a schedule, coordinate runs with an expiring lock, and copy files into controlled category folders.
- Collect and Process Data with a Web Form - validate and normalize public event registrations, store deterministic records, suppress duplicates, and send confirmations.
- Build AI-Powered Issue Triage - normalize issue webhooks, classify untrusted content with a strict AI contract, and gate optional provider write-back behind deterministic policy.
- Build a Messaging Assistant - receive verified provider messages, isolate conversation memory, validate AI drafts, and reply through a deterministic messaging action.
- Build a Queue-Driven Processing Workflow - validate queued jobs, acquire an atomic lease, perform idempotent effects, commit durable completion, and acknowledge only after success.
- Build a Read-Only DevOps Incident Analysis Workflow - collect bounded metrics, logs, and deployment evidence through read-only identities, join separate source branches, and render a human-reviewable incident brief.
Before using a guide
Build and test examples with non-sensitive data in a dedicated project. Provider credentials belong in integrations, General Vault, or private connections, never in templates, block labels, screenshots, or workflow exports.
Most production guides also require you to verify:
- the personal or team ownership scope of every referenced resource;
- available compute, AI, storage, and concurrent-run capacity;
- error branches and alerts for external calls;
- idempotency before sending messages or changing external state;
- the published version and its live trigger, not only the Designer draft.
Related documentation
Explore this folder
Build an Automated Email Response WorkflowBuild a workflow that reads new inbox messages, generates a structured AI response, and sends the validated reply to the original sender.Process Uploaded Files from an EndpointBuild an HTTP workflow that validates an uploaded text file, stores the normalized result, and returns a synchronous JSON response.Advanced File ProcessingBuild a two-path pipeline that stages uploaded files, locks scheduled processing, classifies each file with AI, and stores it by category.Collect and Process Data with a Web FormBuild a web form workflow that validates event registrations, stores deterministic records, and sends confirmation emails to attendees.Build AI-Powered Issue TriageUse an Issue Trigger and an AI Agent to classify incoming Jira, GitHub, or GitLab issues without immediately changing provider state.Build a Messaging AssistantBuild a messaging assistant that preserves conversation identity, generates and validates bounded AI responses, and replies through the source channel.Build a Queue-Driven Processing WorkflowBuild an idempotent queue consumer that validates jobs, controls acknowledgement, handles retries, and routes terminal failures safely.Build a Read-Only DevOps Incident Analysis WorkflowBuild a read-only incident workflow that correlates bounded monitoring, logging, and deployment evidence into a reviewable brief.
