Chat Preferences
What chat preferences do
Chat preferences control how the Cognivo assistant talks to you. You pick the explanation style, how long answers run, whether evidence links appear, and which chains chat leans toward when your question does not name one. Your choices apply to new conversations.
When to use it
Use these settings when the answers you get are not pitched right for you. If chat gives you less detail than you want, switch Response length to Detailed. If it keeps explaining terms you already know, change the default style. If you research one chain most of the time, set it as a preferred chain so you stop repeating it in every question.
Where to find it
Open Settings from your account menu, then choose Chat & AI in the settings navigation on the left. The panel is headed "Chat & AI", with the line "How the Cognivo assistant talks to you."
- Open the Default style dropdown to set the explanation style every new chat starts with.
- Choose Concise or Detailed under Response length for shorter answers or fuller ones.
- Turn on Show evidence links to have on-chain evidence links included in answers.
- Note that Show limitations is marked Always on, so answers keep telling you what a result does and does not cover.
The other rows in the panel
Default style offers Auto, Beginner, Degen, Investor and Forensic. Auto lets Cognivo read your question and choose. The other four hold the same tone for every new chat, from plain first-time explanations through to a forensic write up.
| Setting | What it does |
|---|---|
| Use my first name | Lets the assistant address you by your first name. On by default. |
| Follow-up suggestions | Shows suggested next questions under each answer. On by default. |
| Preferred chains | Biases chat toward the chains you pick when your question does not name one. Ethereum, Base and BNB Chain are available. Left empty, Cognivo chooses the chain it thinks you mean. |
| Personalize Telegram replies | Applies your chat style and personalization to the Cognivo Telegram bot. Off by default. |
What to expect
Each change is saved as you make it. If a change cannot be saved you see "We could not save your change. Please try again." Try it again, and check that you are still signed in.
Style and length shape the wording, not the findings. A short factual answer stays short even on Detailed, because Cognivo will not pad an answer to fill a setting.
Evidence links appear where there is an on-chain record to link to. Some findings come from Cognivo Augmented Intelligence rather than directly from the chain, and those may arrive with no link. That means Cognivo could not point you at a chain record for that specific line, not that the line is wrong. The reply says which kind of evidence it is using.
Preferred chains are a preference, not a filter. Naming a chain in your question overrides the setting, and leaving the field empty is a valid choice rather than a missing one.
Limitations and cost
Changing these settings is free, and chat itself is free. Preferences change how a result is written, never what Cognivo was able to verify.
Critical limitations are always shown and can never be hidden, so you are never misled about what a result covers. That is why the Show limitations toggle is locked.
Three tools charge credits, and preferences do not change that. Bubble Maps costs 1 credit, Smart Accounts costs 2 credits and Wallet Intelligence costs 6 credits. Cognivo quotes the cost first, charges only after you agree, and only when a real result comes back. Every account gets 5 free credits a day, which reset at midnight UTC.
Interface preferences in Cognivo Chat
The Cognivo Chat window has its own display settings, separate from the panel above. Open them from your user name or avatar at the bottom of the left sidebar, then Settings, then the Chat tab.
| Setting | What it does |
|---|---|
| Enter to Send | When on, Enter sends your message and Shift+Enter adds a line break. Turn it off if you write long multi-line messages. |
| Maximize Chat Space | Expands the message area to use more of the screen width. |
| Center Chat Input on Landing | Centres the input field when no conversation is open. |
| Show Scroll Button | Adds a button to jump back to the bottom of a conversation. |
| Font Size and Chat Direction | Set message text size, and left to right or right to left text. |
| Show Code and LaTeX Parsing | Control syntax highlighting for code blocks, and rendering of maths written between $$ markers. |
| Show Thinking | Displays reasoning steps for models that support them. On other models it changes nothing. |
| Save Drafts | Keeps unsent text when you switch conversations, and restores it when you return. |
| Default to Temporary Chat | Starts new conversations as temporary, so they are not saved to your history. |