October 2026

19-Oct-2026 — Actions

Flexivity AI can now make changes as part of a customer’s support request, not only answer it. When a request needs something done in one of your own systems, such as refunding a charge, the AI prepares the change, checks it, and it is carried out by a person or by the AI, as you decide. Progressive Autonomy, which already controlled how far the AI goes in a conversation, now extends to actions.

Four write modes, per tool and per brand — a second ladder beside the seven conversation levels, from the least the AI does to the most:
- Manual — the AI hands the ticket to a person with what to do, and prepares nothing.
- Assist (the default) — the AI prepares the change on the ticket and an agent approves, edits the amount, or rejects it.
- Hold — as Assist, while the customer stays in the conversation: they are told a colleague is confirming, and told it is done once the change is approved and your system confirms it.
- Execute — the AI makes the change itself, within the tool’s limits, and tells the customer once your system confirms it.
The two ladders work together — Hold and Execute apply at the Resolve level, where the AI carries a conversation on after a change; at every other level they work as Assist.
A card on the ticket — every prepared change appears in the Flexivity panel with what it will do, every check it ran (each Passed or Not passed), and when it expires. Approving runs it exactly once; the decision is recorded on the ticket as an internal note.
Checked before it runs, and again when it runs — the record being changed must come from a lookup for the customer on that ticket, every tool has limits (by default at most one change per customer per 24 hours, which you can tighten but never loosen), and a change that moves money needs a largest amount per change before the AI may make it unattended.
Safeguards for unattended changes — only an administrator can choose Execute; a circuit breaker puts a person back in the loop when unattended changes run unusually; agents can mark a change the AI made This change was wrong; and each tool’s page shows what happened to its changes, by brand and mode.
On Zendesk and osTicket tickets — both hosting modes of osTicket included (osTicket needs Flexivity plugin 2.10.0 or later, and a self-hosted install agent 1.6.0 or later).

Docs: Product Overview — Capabilities · Prepared Changes (Zendesk) · Prepared Changes (osTicket)

19-Oct-2026 — Data Connectors

Data Connectors link the AI to the systems that hold your customers’ records, so it can look things up, and make changes, on the customer’s behalf. They are no longer in beta.

Tools that read and tools that change — a lookup, such as a customer’s recent charges, answers a question with the customer’s own records; a change, such as a refund, is carried out as its write mode says (see Actions, above).
Stripe — five tools: find the customer, list their recent charges, look up one charge, list their recent invoices, and refund a charge in full or in part. A charge or invoice belonging to another customer is never returned.
Shopify (beta) — orders, products, and pages.
Any MCP server — connect a server that speaks the Model Context Protocol, add its tools one by one, and choose what each one does. Zapier is supported through its MCP server, which reaches 9,000+ apps: a Google Sheet, a Zapier Table, or an app your team already uses.
Always about the customer on the ticket — Flexivity takes who the customer is from your help desk. Nothing the AI writes, and nothing typed in a conversation, can point a lookup at someone else. In chat, a customer is looked up only when they are signed in.
You choose what the AI sees — for each tool, each field of the result is passed to the AI or withheld. Flexivity keeps little of what it reads, and never a name, address, email address or phone number.
Nothing runs until you enable it — every tool starts Off, can be tested first (including as a ticket’s own requester), and each connector shows its health and recent changes.

Docs: Data Connectors

19-Oct-2026 — Procedures

Procedures tell the AI how to handle a kind of request, in plain language: what to look up, what to tell the customer, what to ask, what to change, and when to hand over to a person. They bring conversation logic and your Data Connector tools together, without code.

Three ways to start — from one of five samples (Order status, Cancel before shipping, Damaged item refund, Damaged item refund (store order) and Address change), by pasting the procedure your team already follows, which Flexivity drafts for the AI, or from a blank page.
Name a tool by typing @ — or write “look up the order” and Flexivity proposes the right tool for you to confirm.
See what Flexivity understood — before a procedure goes live, the Reading tab shows its steps in order, the tool each step uses, the questions it may ask (including which ones ask for a file, such as a photo of the damage), its conditions and its hand-overs.
Code enforces the steps that matter — a change is prepared only when the procedure reaches it, its target must come from a lookup for this customer, and a condition such as “only if the order has not shipped” is checked against that record before the change can run.
Test conversations — check a procedure behaves as you expect, including that a change is not prepared when it should not be. The samples that change something come with one of each.
Versions — editing creates a draft and the published version keeps running until you publish it; a procedure that changes something is published only once its tests pass, and each conversation keeps the version it started with.
Per brand — a brand also uses its parent’s procedures unless it has its own with the same name.

Docs: Procedures