C
Crestline / ai
02WorkflowsLive

Describe it. Watch it run.

Type the automation in plain English and a five-agent pipeline turns it into a validated, running workflow grounded in your org graph.

For COO · OperationsLayers on n8n / Zapier — org-native
Overview

Classic automation tools are great at the predictable but useless for the unstructured work — a decision in a thread, a task that just got blocked, a new hire who joined yesterday. Crestline takes a sentence and builds the automation: intent, planner, drafter, validator and narrator run in sequence, and you watch each agent finish before activating the draft.

What makes it more than a prettier Zapier is that the building blocks are native to your organisation. Triggers fire on real events; steps create tasks, write to memory, notify people and log decisions; and nothing stores a raw email or ID — org-graph tokens like @manager_of resolve at run time, so automations survive a reorg.

What you get
Five-agent builder

Intent → Planner → Drafter → Validator → Narrator turns a plain-English request into typed, schema-validated workflow JSON.

Org-graph tokens

Route to @manager_of(@self), @dept-finance or @cxo — resolved live, so workflows keep working when the org changes.

Native triggers & steps

Fire on a task assigned, a status flipped to blocked, a mention, a schedule or a webhook; create tasks, reply, notify or log a decision.

Visual graph + full audit

Every run keeps a per-step trail; the spec renders as a directed graph with live run status overlaid.