Workflows & agents

The routine, dispatched — under your rules.

Rules-based workflows for the predictable work. AI agents for the open-ended work. Both run on the same data model, under the same controls, with the same audit trail.

modulaw / workflows

60%

Less time on admin

24/7

Scheduled workflow runs

Audit

Trail on every action

Yours

Rules, your control

Capabilities

What’s inside.

Workflow builder

Compose triggers, conditions, and actions visually. Document generation, intake routing, deadline reminders — without code.

AI agents

AI agents that draft, research, follow up, and progress matters on a schedule — within firm-defined rules and controls.

Triggers

Time-based, event-based, document-based, or matter-status-based. Workflows fire when the right thing happens.

Approvals

Human-in-the-loop where you want it. Critical actions surface for partner review before execution.

Templates

Pre-built workflows for engagement letters, conflict checks, intake triage, hearing reminders, and invoice cadences.

Agent skills

Configure what each agent can and cannot do. Skills are scoped per agent and audited per use.

Versioning and rollback

Every workflow and agent has a version history. Revert a change with one click.

Audit trail

Every workflow run and every agent action is logged with timestamps, inputs, outputs, and the human who reviewed it.

In depth

A closer look.

Two modes

Rules-based when the work is predictable. Agentic when it isn't.

Use workflows for the work you can fully describe — intake triage, deadline reminders, follow-up cadences. Use agents for the work that needs judgement — drafting, research, correspondence. Both share the same data and the same audit infrastructure.

  • Workflows: deterministic, fast, ideal for compliance-sensitive tasks
  • Agents: contextual, exploratory, ideal for drafting and research
  • Switch between modes mid-flow when the work requires it

Control

AI you can actually deploy.

Every agent runs under explicit rules. Every action is logged. Every output is reviewable. Nothing happens that you can't see, reverse, or attribute. That's the difference between AI you can demo and AI you can deploy at a firm.

FAQ

Questions, addressed.

Can AI agents act without review?

Only when you configure them to. Critical actions can require human approval; routine actions can run on a schedule under your rules. You set the threshold per skill.

Is everything an agent does logged?

Yes. Every input, output, decision, and tool call is logged with timestamps. Audit logs are exportable and tamper-evident.

How do workflows differ from agents?

Workflows execute deterministic rules — same input, same output. Agents reason about open-ended work. Both share the same data layer.

Are the workflows hard to build?

No. The builder is visual and most firms start with the bundled templates, then customise. No engineer is required to get value on day one.

See it against your matters.

Start free in minutes, or book a demo and we’ll run it on your firm’s real work.