Skip to main content

Copilot

The Copilot is the most useful day-to-day tool for working with your AI flows. Instead of editing forms by hand, you talk to an assistant that can read every configured flow, explain how things work, make changes for you, and troubleshoot when something misbehaves.

It's under active development, so expect occasional rough edges — but it's the primary way the team builds and maintains everything from flows and data points to integrations, tickets, WhatsApp templates, and FAQs.

Features​

The Copilot's capabilities group into a handful of areas. Each has its own page with sample prompts you can adapt.

AreaWhat it handles
Understand & planExplain configuration, review a flow, or plan an approach — read-only.
Build & change flowsFlows, objectives, branching, data points, and child-ticket templates.
Connect external systemsIntegrations, custom APIs, source APIs, and sample data.
Tickets & dispositionThe disposition form, on-closure actions, and per-flow presets.
WhatsApp & outreachWhatsApp Flows, broadcast templates, and template reply hooks.
Knowledge & repliesFAQs, agent-facing canned responses, and account policies.
PoliciesBusiness rules linked to your configuration, checked when it changes.
TestingDraft test cases and review why a test passed or failed.
DebuggingWhy a flow, transition, custom API, or ticket misbehaved.

Copilot layout: Chat, Split, and Workspace​

The Copilot has two panes: the chat with the assistant, and the workspace with your configuration (Flows, Data Point, Custom API, FAQ, Policies, Ticket Fields, Library, Issues, Tests) and the tools Test AI, Manual Ticket, and API Playground. The switch next to the Copilot title picks how they share the screen; Ctrl+\ (⌘+\ on Mac) cycles through the three.

  • Chat shows only the conversation, at a comfortable reading width. A new chat opens here. On wider screens, a sidebar on the left lists your chats by date (Today, Yesterday, Previous 7 days, and so on), with search and New chat; click a chat to open it in place. Hide the sidebar with the button next to New chat, and the Copilot remembers that choice. In Split and Workspace, chats open in tabs above the conversation instead, and History lists them.
  • Split puts the workspace on the left and the chat on the right. Drag the divider to resize; double-click it to go back to half and half. Use it to debug: open a conversation in Test AI on the left and ask the Copilot about it on the right.
  • Workspace gives the workspace the full width, for tables such as Tests and Issues. Ask Copilot in the corner brings the chat back beside it.

You can click New chat while the current conversation is loading or Copilot is responding. The previous chat keeps loading or working in the background while you start a separate conversation.

The Copilot remembers the layout and divider position you pick, in this browser. Some actions change the layout for you: clicking a flow or data point in the chat opens it in Split, and debugging an issue or a conversation opens in Split. On small screens, Split shows the chat, and Chat and Workspace buttons switch between the two.

Read a flow in the Copilot Flows tab​

When conversation audits or automation gap analysis are enabled, Needs attention appears above the starter tips in an empty Copilot chat and highlights up to three concerns; it hides once the chat has messages. The list considers priority, affected-ticket volume, task usage, and issues observed after a previous resolution. Each flow's detail card also shows up to three concerns for that flow.

Click a concern to see its first known and latest occurrence, affected tickets and customers, task versions, previous resolutions, and source evidence. Review fix with Copilot opens the evidence and history in a new Copilot chat for preparing a change. Opening a concern does not mark it fixed or remove it from the open queue. Review the proposed change and test it before publishing.

Ticket-usage figures cover tickets opened in the last 30 days that used the flow. The audited count shows how many of those tickets have a processed conversation audit. Observed issues do not establish the failure rate of unaudited tickets. Missing dates and versions stay unknown; versions reconstructed from history are labeled inferred.

If evidence arrives after an issue was marked resolved, its occurrence date determines whether it is historical evidence or a later observation. Later observations distinguish older task versions, the version recorded at resolution, and possible regressions in later versions. A captured occurrence on the resolved version or a later version reopens the issue. An uncertain version is flagged for verification; an older version still being used is shown separately. After reviewing or fixing a concern, Mark resolved records the current published task version. A resolution records the configuration at review time, rather than proving that a fix worked.

Each flow in the Copilot's Flows tab has a one-line header: its name, Entry if it starts conversations, a short summary such as "11 objectives · 4 routes · hands off to an agent", and its status: Live, Draft (with publish, compare, and discard buttons), Not published, or Off. When a flow is open, its header stays at the top while you scroll, with buttons for the flow graph, Version history, and a ⋯ menu with Force sync to the live flow.

An open flow shows:

  • Starts when: the trigger condition, and which flows route here. The summary only appears when it says something the trigger condition doesn't.
  • Objectives: one row each, with the wording the customer or agent sees and where its routes lead. Each row is colour-coded by kind, on its left edge and its tag: purple for an execute action, cyan for a data point, green for computed, blue for asks customer, and orange for asks agent. Click a row for its full settings, or the gear to edit it.
  • Finishes with: the completion prompt, the team, and whether an agent takes over.
  • Routes, in order, then data points, disposition presets, notes to AI, and task settings.

In the Workspace layout on a wide screen, the flow details sit in a column on the left and the objectives take the main column on the right.

When you debug a conversation or an issue, the flow it was in opens with a summary at the top, such as "Waiting on cancellation_reason" or "Routed to delay_delivery", with counts of collected, skipped, and remaining objectives. Each objective row shows its value from the conversation and its state: collected, next, to ask, skipped (with the reason), or not reached. The Copilot scrolls to the objective the conversation is waiting on. Run details lists the next step, the matched route, and the flows the conversation passed through.

Business Opening Message​

Ask Copilot to set the business opening sentence and configure its quick replies, including buttons that start specific tasks:

"Set our opening sentence to 'Welcome to Acme! How can we help?' and add a 'Track order' button that opens the Order tracking task."

You can also ask to view the current greeting and button mappings, rename or remove a button, change its task, or clear the greeting. Changing only the greeting leaves the buttons intact. Opening quick replies support up to 10 buttons with labels of up to 20 characters. Task mappings use saved, enabled tasks in your account; Copilot asks for clarification if a requested task is missing or ambiguous.

To create a task and a welcome button together, include both in your request. Copilot saves the task before configuring the button that opens it.

To include each customer's name, ask Copilot: Set our opening sentence to 'Hi {{customerName}}, Welcome to Acme!' The placeholder uses the contact name, or “there” if no name is available. Personalization is optional: leave out {{customerName}} for a fixed greeting. See Personalize the Opening Sentence.

How a Request Is Handled​

Working with the Copilot follows the same loop no matter what you're changing:

  1. You describe what you want in plain language — a question, a change, or a whole new flow.
  2. The Copilot reads your live configuration — every flow, data point, custom API, template, and FAQ in the account — so its answer or change is grounded in what you actually have, not a generic template.
  3. It plans the work. For anything non-trivial it breaks the request into ordered steps and streams its progress as it goes, so you can see what it's doing and why.
  4. It asks before guessing. If something is ambiguous or a required detail is missing, it stops and asks a short question (usually with suggested options) rather than inventing an answer.
  5. You review the result. Read-only requests just answer. Changes are made against your configuration for you to review — then confirm they behave with a Test flow run or an automated test.

The clearer your request, the fewer round-trips: name the customer intent, the specific flow or data point, and the outcome you want.

When a Request Spans Several Areas​

You can ask for several things at once and the Copilot will break the work into ordered steps, respecting dependencies (for example, creating a data point before the flow that uses it).

"Add a data point for tracking and update the refund flow to use it." "Create a refund flow and add its disposition presets."

If you paste mixed content — say, some fixed policy facts plus flow-style instructions — it splits that into FAQs and flows so nothing gets dropped.

When It Asks You to Clarify​

  • If a request is too vague to act on confidently, the Copilot asks a short clarifying question with suggested options rather than guessing.
  • If a request is outside what it can configure (e.g. product UI changes, analytics/report generation, billing, or editing source code), it tells you and points to the closest supported alternative.

Long Conversations​

The Copilot keeps the messages of a chat in its context so it can refer back to earlier work. When a thread grows large, it automatically summarizes the older messages and keeps only the recent ones verbatim. A divider reading Earlier messages summarized marks where this happened in the transcript.

Nothing is deleted. The earlier messages stay visible, and the summary preserves the flows, data points, APIs, decisions, and open questions they contained, so you can keep saying "edit that flow" without repeating yourself. If you want a completely fresh context, clear the chat or start a new one; clearing a chat also discards its summary.

Use Copilot from Other AI Apps​

Copilot also works from AI apps that support MCP connectors, such as Claude, ChatGPT and Cursor. Add https://api.flowcall.co/mcp as a connector, authorize it in Flowcall, and the app can hold Copilot conversations, check the saved configuration, and save or publish task changes when you ask. It acts as you, and its conversations appear in your Copilot history. See Use Copilot from Claude, Cursor and other AI apps.

Tips​

  • Use it to learn an unfamiliar account before changing anything — ask how existing journeys work.
  • Be specific about the customer, the trigger, and the desired outcome.
  • Name exactly what you mean when it matters ("edit the data point, not the API") to avoid ambiguity.
  • After the Copilot makes a change, validate it with the Test flow feature or an automated test run.