Every agent in eChat has a handful of settings that shape how it sounds, what it knows, and how it works with your team. These live under screens in the agent editor: Instructions, Product, and Agent.
None of these are set-once-and-forget. As your product changes, your training material grows, or you notice the agent handling something oddly, you come back to these screens to adjust it.
Instructions
What it is
The Instructions field is a plain-text box where you tell the agent how to behave: its tone, what it should focus on, and what it should never do. It is the closest thing to giving the agent a personality and a set of rules.
What it's best for
- Setting a tone — casual and friendly, or formal and precise
- Steering the agent toward certain topics and away from others
- Hard rules, such as never quoting a price without a disclaimer
- Handling edge cases you've noticed in real conversations
- On eWebinar agents, telling the agent when to stay silent on a poll or interaction-card answer (see below)
How it works
When you first set up an agent, the setup wizard asks you to pick a Role — Sales (the default), Customer Success, or Marketing — and seeds Instructions with a starting set matched to that role. If you switch roles after editing the instructions, eChat asks before replacing your edits. The agent treats these instructions as standing guidance for every reply, on top of whatever it pulls from its training material.
Staying quiet on card answers. In an eWebinar, your Behavioral instructions can tell the agent when not to reply to a poll, quiz, or question-card answer. Typed chat messages are always answered. See how the AI answers.
Best practices
- Write specific rules, not vague ones — "never mention pricing under $499" works, "be helpful" doesn't
- Update instructions after you spot a bad reply in Chat Logs
- Keep it short enough to scan
- Don't cram product facts in here — that's what training material and the Product summary are for
Product
What it is
The Product screen holds the Product name, the Language the agent works in, and an AI-generated summary of what your agent is talking about. eChat builds the summary from the home page at setup and from training material after the first crawl.
What it's best for
- Giving the agent a clear, condensed sense of what it's representing
- Spotting when training material is vague or scattered
- Keeping an agent focused if its training touches on adjacent topics
How it works
eChat drafts your product summary for you, first from your home page at setup and again when your first crawl finishes. Your edits are always kept, and Keep & edit current version keeps a draft you like. Whenever you want a fresh pass, use Re-generate Product Summary from Training Data, or edit the text by hand. One agent should stick to one product — create a separate agent per product if you sell more than one.
Best practices
- Regenerate the summary after a significant training update
- Read it after every regenerate — if it drifts, edit it directly
- Treat a confusing Product summary as a signal to clean up training material
- If you like the home-page draft, choose Keep & edit current version before the first crawl finishes
Agent
What it is
The Agent screen sets the agent's identity — its name, avatar, and whether it's labeled as AI — along with which AI model it uses, how strictly it checks its answers against training material, and how it works with your team under Escalation.
What it's best for
- Making the agent feel like a real member of your team in chat
- Balancing reply quality against cost by choosing Auto (Smart), Auto (Efficient), or a specific model
- Tightening or loosening how carefully the agent double-checks itself
- Deciding whether the agent stays alone, leaves notes for your team, or can bring a person in
How it works
Identity. Set a Name and an avatar (Upload image, or Replace image). In an eWebinar webinar, this is what attendees see in the live chat. Turn on Show as AI to add an "(AI)" badge after the agent's name; the label syncs to eWebinar when linked.
AI Model. Choose how the model for visitor replies is picked. The picker lists two Auto settings first, under Recommended: Auto (Smart), the default, and Auto (Efficient), the lower-cost one. With either, eChat picks the best model for you and keeps it current. You can also pick a specific model. See Choosing an AI Model.
Answer confidence. Set Grounding strictness to Creative, Balanced (the default), or Strict. Stricter settings mean more checking against training material. At every setting the agent answers in its own words.
Escalation. Escalation is three fixed steps: whether the agent involves a person, when, and how. Step 1 is:
- Agent only — works on its own.
- Leave a note — can flag messages as Needs reply and leave private notes, while it keeps helping the visitor. When it flags for your moderators, that can email them so they see the note.
- Bring in a person (or Bring in a moderator, for eWebinar agents) — everything Leave a note does, plus the agent can hand a conversation to a person when you've allowed it.
With Bring in a person, the when cards decide which situations can lead to involving someone (for example uncertain answers, frustrated language, or an explicit human request). For eWebinar agents, how is your eWebinar moderators. Whether a hand-off is possible on a given webinar is controlled in eWebinar. See Working with Human Moderators.
Wait time (set in eWebinar). How long the AI waits for a human to answer first isn't an eChat setting — it's AI answers if a human doesn't answer in … sec in each webinar's AI moderator settings in eWebinar (up to 15 minutes). See the wait time.
Best practices
- Start with Auto (Smart) and Balanced grounding
- Move to Strict grounding for anything regulated or high-stakes
- Use Creative sparingly when training material is thin
- Pick a name and avatar that read as a real person or team identity, and use Show as AI if you want to be upfront
- Only choose Bring in a person if someone actually watches the chat; otherwise Leave a note gives your team a Needs reply list without promising live help
- If your team moderates live, keep a short wait in eWebinar so people answer first; for sessions nobody watches, set it to 0
Welcome Message and Visitor Settings
What it is
eChat also has Welcome Message and Visitor Settings screens (a greeting, and name and email collection). They belong to standalone chat agents, so they don't appear in the editor for agents linked to eWebinar.
How it works
For eWebinar-linked agents, eWebinar already knows who each attendee is and runs the chat room, so the agent doesn't send its own greeting or ask attendees for their name and email. If you're looking for these screens on an eWebinar agent, that's expected - there's nothing to configure.
Best practices
Put anything you'd have said in a greeting into Instructions instead (for example, what the agent can help with), and manage attendee details in eWebinar.