Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/arainey2022/myaskai-docs/llms.txt

Use this file to discover all available pages before exploring further.

The most common question new customers ask before signing up is: “How long will this actually take to set up?” The answer is that most teams have their AI agent writing draft replies to real tickets on the very first day they sign up. Getting to fully autonomous, customer-facing replies typically takes a few days to a few weeks, depending on how much testing and refinement you want to do before going live. This guide gives you a clear picture of what to expect — and what to do — at every stage of the journey.

​
What you need before getting started

No specialist skills or technical knowledge are required to set up My AskAI. Here’s everything you need:
  • A company email address — personal email addresses are not accepted at sign-up
  • Admin access to your helpdesk — required to authorise the My AskAI integration (Intercom, Zendesk, Freshdesk, HubSpot, Gorgias, etc.)
  • Existing help documentation or a website — My AskAI trains your agent on content you already have; the more comprehensive your documentation, the better your agent’s answers
Developer involvement is only needed if you later choose to connect the User Data API (for personalised responses) or set up AI Actions (Tools) for back-end integrations. Core setup is entirely no-code.

​
Your onboarding checklist

Work through these tasks in order. Most teams complete the core setup in under 30 minutes.
1

Complete initial setup and connect your knowledge base

Follow the Create Your AI Agent guide to sign up, add your first content source, and connect your helpdesk. By the end of this step your agent will be generating AI-suggested replies as internal notes on incoming tickets — visible to your team, not your customers.Quick-start options:
  • Connect a Help Center (Intercom, Zendesk, Freshdesk, GitBook) for automatic syncing
  • Enter your website URL for a full site crawl
  • Upload PDFs or documents directly
Help Center connections are the highest-quality content source and sync on a schedule — always prefer a direct connection over a website crawl if it’s available for your platform.
2

Configure guidance and tone of voice

Visit the Guidance section of your dashboard and set the ground rules for how your agent communicates. Useful starting points:
  • Define your brand’s tone (professional, friendly, concise, or detailed)
  • Set scenario-based rules (e.g. “Always escalate if the customer mentions a refund or legal issue”)
  • Specify any topics the agent should never attempt to answer independently
  • Give your agent a name that fits your brand
Good guidance has a bigger impact on answer quality than adding more content. Spend 10–15 minutes here on Day 1 and refine as you go.
3

Set up human escalation

Configure My AskAI’s human escalation feature so customers who need a real person can reach one smoothly. Set up:
  • Escalation triggers — the conditions under which your agent should offer a handoff to a human
  • Escalation data collection — capture customer name, email, and issue description before the handoff so agents have full context
  • Routing rules within your helpdesk — assign escalated conversations to the right team or agent
Do not skip this step. A well-configured escalation path is essential for customer trust. Customers should always have a clear route to a human agent, even when the AI is handling the majority of queries.
4

Customise the chat widget appearance

Head to the Customisation settings in your dashboard to match the widget to your brand:
  • Set your brand colours and widget button style
  • Upload your company logo or agent avatar
  • Write a welcome message that reflects your brand voice
  • Configure the widget position (bottom-right, bottom-left, etc.)
A well-branded widget builds customer confidence and increases engagement. Customers are more likely to interact with an agent that clearly belongs to the product they’re already using.
5

Connect a live chat channel

If your team uses a live chat platform (Intercom, Zendesk Messaging, Freshchat, HubSpot, etc.), connect it via the Channels section of your dashboard. This enables your AI agent to respond within your existing live chat interface rather than as a standalone widget.Once connected you can also:
  • Enable auto-tagging to categorise conversations automatically in your helpdesk
  • Set up routing logic so AI-handled conversations are triaged separately from escalated ones
  • Install the Chrome Extension so your human agents can use the AI copilot anywhere they work
The Chrome Extension is especially useful during the early days — it lets your support team see AI-suggested answers in their browser without switching between tabs, making it easy to validate quality before fully enabling autonomous replies.
6

Invite your team members

Go to Team Access in your dashboard and invite your support team, team leads, and any stakeholders who should have visibility into the AI agent’s performance.My AskAI supports multiple roles — you can grant full admin access to team leads and read-only access to others who just need to monitor the insights dashboard. Shared visibility helps the whole team align on what the AI is doing and flag issues quickly.
7

Review your first insights report

After your agent has handled its first batch of real conversations, open the Insights dashboard. You’ll see:
  • Top topics — what customers are asking about most
  • Knowledge gaps — questions the agent couldn’t answer (opportunities to add content)
  • Sentiment and CSAT trends — how customers are rating interactions
  • Feature requests, bug reports, and feedback — categorised automatically from conversation content
Use this data to prioritise your next round of content additions and guidance refinements. The insights dashboard becomes more powerful the more conversations your agent handles.

​
Day-by-day timeline: your first 30 days

Day 1 (15–30 mins)

Connect your knowledge base and helpdesk. Add initial guidance. Enable auto-tagging. Invite team members. Your agent starts generating note replies immediately.

Days 2–7 (15–30 mins/day)

Review AI replies in your helpdesk and in Inspect & Logs. Identify knowledge gaps. Add custom answers for high-frequency misses. Refine guidance. Install the Chrome Extension across your team.

Days 8–14 (15 mins/day)

Review Self-learning article suggestions. Enable direct customer-facing replies. Customise helpdesk rules (assignment, tagging, triage) to fit your workflow.

Days 15–29 (15–45 mins/day)

Monitor direct replies. Use Insights to identify low AI-resolution topics and CSAT dips. Build multi-step Tasks for high-impact support processes.
Day 30: Add your payment card and upgrade to a paid plan to continue beyond the free trial. Pro plan customers are largely self-serve; Enterprise plan customers receive white-glove onboarding support including helpdesk setup assistance and a dedicated Slack channel with the My AskAI team.

​
Support during onboarding

  • Pro plan — self-serve with live chat support available for helpdesk connection issues or bug reports
  • Enterprise plan — white-glove onboarding including a My AskAI team member with access to your helpdesk, setup calls, and ongoing Slack support throughout onboarding and beyond
If you have questions at any point, the My AskAI support agent is available in the chat widget on every page of this documentation.

Build docs developers (and LLMs) love

Get started for freeTalk to us