The Agentic Workflow pattern

In one line. A multi-step automation that processes work and decides — documents in, decisions out, humans approving only the edge cases. You’ll be able to. Recognize workflow-shaped problems, assemble the eight-part processing loop, and choose the right engine for the middle of it.

What it is

An Agentic Workflow is level 2 of the ladder. Where Search answers questions, a Workflow does the work: it takes each incoming document through extraction, validation and decision, routes the uncertain minority to people, and shows everyone the state of everything.

Choose this shape when the sentence in your head is “we process these every day and it’s manual” — invoices, claims, contracts, applications, tickets — and there is a decision at the end: settle, escalate, approve, route.

The processing loop

Same spine as Search (collection → ingest → agent), plus a lifecycle, a human, and an operations surface:

1. PROCESSING collection — the container
2. SCHEMA                — the fields to fill
3. LIFECYCLE             — the stages, and the gates between them
4. INGESTION             — upload or connected source
5. ENGINE                — agent / team / XFlow / mesh, set as the collection's intake agent
        │ each document walks the lifecycle
        ├── 6. HUMAN-IN-THE-LOOP  — exceptions reach the Inbox; a person decides
        ├── 7. OUTPUT / EXPORT    — views, datasheet, export template
        └── 8. EXPERIENCE         — lists, inbox, operations pages for your team

The build order matters less than the design order: schema and lifecycle first. They are the contract everything else fills.

1. Processing collection — Studio ▸ Data ▸ Collections ▸ + Collection, and set Purpose to Processing. (D1)
2. Schema — the fields a settled record must have: vendor, amount, po_match… (D1)
3. Lifecycle — the stages a document moves through, and the automations that move it: New → Extracted → In review → Settled. Confidence gates live here — this is where “humans only see the edge cases” is actually wired. Open the collection and use its Lifecycle tab; Lifecycle, Taxonomy, Tags and Events all live inside the collection, so don’t hunt for them in the list on the left. (D3)
4. Ingestion — how the work arrives: upload documents by hand, or connect a source (SharePoint, email, a drive, a feed, blob storage) and schedule the pull under Studio ▸ Data ▸ Ingestion. (D4)
5. Engine — build the worker, then open the collection’s Agents tab (Studio ▸ Data ▸ Collections ▸ your collection) and set it as the collection’s intake agent, so it runs on every document that arrives. One agent for simple extraction; a team or mesh when steps deserve their own specialists; an XFlow when you need scheduled or branching pipelines.
6. Human-in-the-loop — route low-confidence or flagged documents to the Inbox; reviewers correct, approve, and their corrections become feedback.
7. Output — views (Studio ▸ Data ▸ Views) and a Datasheet for downstream consumption, export templates (Studio ▸ Data ▸ Export) for other systems.
8. Experience — the workspace is live from day one; when you want an operations view, design a page under Studio ▸ Experience ▸ Pages and drop in cards for throughput, exceptions and aging.

Choosing the engine

Engine	Use when	Chapter
Single agent	One coherent job — extract and flag	A3
Team / mesh	Distinct steps deserve distinct specialists: classify → validate → enrich → approve → summarize, each lighting up live	A11
XFlow	Scheduled runs, fan-out, branching, cross-collection pipelines	A10

All of them are built on the same page — Studio ▸ Agents ▸ Agents. Flows, teams and meshes appear in that list alongside plain agents; use the list’s filter to see just the kind you want. There is no separate flow or multi-agent entry in the menu.

Invoice settlement is the smallest complete example: invoices land, one extractor agent fills the schema, the confidence gate settles the clean ones and sends the rest to a reviewer, and the document’s Summary pane fills in field by field as the agent works. Loan underwriting, claims adjudication and contract review are the same shape wearing different schemas — with more specialists in the middle.

Worked builds, easiest first

Build	Teaches
B3 · Invoice settlement	The whole loop, smallest form — do this one first
B4 · Support intake triage	Routing, tags, inbox discipline
B5 · Contract review	Clause extraction, four-eyes review
B6 · SEC filings analyst	Scheduled XFlow pipelines over a public feed
B7 · Loan underwriting	Mesh engine, multi-collection, straight-through-processing gate

Do B3 even if your real target is B7. Every workflow is the same skeleton wearing different clothes, and the skeleton is easiest to see on the small one.

Where to go next

• Start building: B3 · Invoice settlement.
• The gates that keep humans in command: G2 · Human-in-the-loop.
• Outgrowing one workflow: the System pattern.