Skip to Content
AgentsBuilding workflows

Building workflows

A workflow is built on its harness: the canvas showing the graph, laid out as Triggers → Steps → Result.

The harness canvas, with an event trigger, work steps, and a wait

A workflow is inert until you enable it, so everything on this page is safe to experiment with.

Creating a workflow

Click New workflow and give it a name. You get an empty draft. Add a summary from the line under the title — Alma uses it, and so does the next person who opens the workflow.

Triggers

A trigger starts a run. Two kinds:

  • Schedule — a cron expression, like 0 9 * * 1 for 9am every Monday.
  • Event — something happened in a connected tool.

Every workflow can also be run manually, at any time, regardless of its triggers.

Add trigger creates an event trigger. To create a scheduled one, add a trigger and switch its type in the editor.

Steps

Each step has a kind:

BadgeMeans
WorkDoes something
ApprovalA person decides
WaitPauses for a condition
BranchThe path splits

And an actor: Alma, a Person, a System, or something External.

Editing a step

FieldNotes
Step nameRequired
InstructionsWhat Alma should do at this step
ModelWorkflow default, or pin to Sonnet 4.6, Opus 4.6, or Haiku 4.5
Require approval after this stepSee below
Retries on failure0–10

Pin a model deliberately: use the workflow default for most steps, and reserve Opus for steps that draft customer-facing content.

Approval checkpoints

Turn on Require approval after this step to stop the run there until a person signs off. Choose who:

  • Anyone on the team — the first person to look at it can approve
  • Specific people — only they can

Write gate instructions telling the reviewer what to check. “Confirm the discount is within guardrail and the usage trend is real” gets a better decision, faster, than “Approve?”

Only people can approve — Alma’s own agent memberships are excluded from the approver list.

Put checkpoints where judgment changes the outcome: anything irreversible, customer-facing, or past a threshold. An approval on a step nobody reviews adds latency without adding safety.

Editing by chat

The rail on the right of the harness edits the canvas by conversation. Describe a change and Alma applies it card by card. Click a card first to give it context.

The harness chat rail, editing the canvas by conversation

Access

The Access tab controls two things.

A workflow's Access tab — tool modes and visibility rules

Tool modes

Every tool a step uses gets a row, set to Automatic or Needs approval.

A tool’s mode applies to the whole workflow, not one step. Changing it rewrites the setting on every step that uses the tool. If any step routes the tool through a person, the row reads Needs approval.

New workflows default to Needs approval, unless your workspace changed that default in Security.

Access rules from Connections still apply on top of this. If a rule denies a tool to the people behind this workflow, setting it to Automatic doesn’t override that — the rule wins.

Visibility

A baseline audience, plus rules for specific groups and people, each with Can view, Can edit, or Full access, as Allow or Deny. Resolution: deny wins, and person > group > tier.

Use the Who sees it now card to check the result before relying on it.

Coming soon. The baseline offers Everyone, Admins & up, and Owners only, but Owners only saves the same as Admins & up and reads back that way. Use a deny rule if you need owner-only access.

Changing the baseline is owner-only. Making a workflow visible to the whole workspace requires org admin.

Saving

Triggers and steps save as a whole list — a deleted step is deleted for real on save. Removing a step’s checkpoint cancels any decision pending on it, so a suspended run aborts rather than waiting on a gate that no longer exists.

Result

The rightmost column shows where runs land: Success, Partial, Cancelled, or Escalated.

Coming soon. Results are shown but can’t be edited on the canvas yet. Ask Alma in the chat rail to change them.

Next

Last updated on