Skip to main content
This changelog lists dated updates to the Fini (usefini.com) documentation and product, newest first. Each entry says what changed and links to the affected pages.
Agent Handshake: one answer for customers' personal AI agents

Agent Handshake

Customers now send personal AI agents, such as Meta’s Muse, OpenAI’s Dots and Google’s Gemini Spark and Call for Me, to handle refunds, cancellations and billing questions. Agent Handshake makes sure the same request gets the same answer, whoever is asking and however many times they ask.
  • Same request, same decision. Fini decides from the customer’s intent, live account data and your written rules, so a reworded request reaches the same rule.
  • Every reply is checked. Guardrails check for banned terms, confidential details, allowed links and your plain-language rules. A reply that fails is rewritten or handed to your team, and flagged conversations show in Inbox.
  • It knows when to stop. Fini escalates to your team, or sends the agent back to its owner, when a request needs a person.
  • You stay in control. Likely personal agents are tagged in the Inbox, and you choose which actions an agent can ask for and which need approval.
Agent Handshake is open to everyone: any team running an AI agent for support, ours or not. On Fini it is switched on per account. See Handle customers’ personal AI agents, the Agent Handshake page and the launch post.
Scout: an AI coworker for support teams, in Slack

Scout

Scout is a new Fini product that works in your support team’s Slack. Ask it questions in plain language and it reads across your tickets, reviews, threads and call transcripts, finds the patterns, and drafts the follow-up: help articles, bug reports for Jira or Linear with the root cause attached, and notes to product in the customer’s own words. Quick questions are answered in seconds, deep ones in minutes.You choose what Scout watches: every conversation, or only the checks you set up. It flags accounts that show churn risk, and the Monday report includes a QA summary of your human agents alongside QA of your AI agent. Scout is self-learning: corrections you give it carry into every later answer and report.New connections are read-only by default. Any change Scout proposes waits for a named person to approve it, and every run is logged. Scout works with any helpdesk and any customer-facing agent.Scout is available now. See the Scout docs for setup, safety and connectors, and get Scout in your Slack; our team will set it up with you.
Documentation: new Trust & pricing and Use cases tabs, Performance & accuracy group, playbooks and new guides

New tabs

  • Trust & pricing: a new tab covering security and compliance, plans and billing, and material for evaluating Fini, including What counts as a resolution and Security overview.
  • Use cases: a new tab that brings together use-case guides (refunds, disputes, cancellations, escalation and handoff, KYC and account changes, billing and invoices, password reset), the existing walkthroughs, industry setup for fintech and healthcare, and a new How-tos group.

Performance & accuracy

A new group under Documentation that explains how to read your agent’s results:

New Getting Started pages

New guides

Playbooks

Seven vendor-neutral playbooks in the Use cases tab: escalation policy, preparing your APIs, testing before launch, running a pilot, knowledge base prep, multilingual support, and migrating from a legacy chatbot or IVR.

Updated integration pages