Search & Chat · June 12, 2026

One question, all your notes as context

Imagine having an assistant that read all your notes, receipts, contacts and reminders, and can answer any question based on that information, and still shows you where each answer came from.

That’s exactly what CleverNote’s chat does. You ask, it searches, crosses references and answers with citations you can click to see the original note.

What the chat can answer

The chat has access to five internal “tools” it uses automatically based on the question:

Note search: any question about text content. “What did I note about project X?” “Is there a note with my mom’s WiFi password?”

Entity discovery: finds a person, company or service provider by need or category when you don’t know the name. “I need a plumber, do I know one?” “Who recommended the nutritionist?”

Memory focused on a person or provider: when you ask about a specific person mentioned by name, it searches only that entity’s notes and brings up their contacts. “What’s my dentist’s phone number?” “What did Bruno say in the meeting?”

Financial query: sums your expenses by category and period, citing each line item. “How much did I spend on restaurants in April?” “What’s my biggest expense this quarter?”

Reminders: dates and upcoming appointments. “What do I have scheduled this week?” “When was the last reminder about insurance renewal?”

For questions that cross domains, the chat uses multiple tools and combines the answers.

How citations work

Each part of the answer that comes from a specific note appears with a clickable link. Clicking goes directly to the source note. This serves two purposes:

1. Verification: you can check if the answer is correct
2. Additional context: the original note has more details than the answer summary

If the AI drew an incorrect conclusion, you see the source and can correct both the answer and the underlying note.

Examples of questions that work well

What was the last quote I received for bathroom renovation?
→ Cites the meeting note with the contractor and mentioned amount

How much did I spend on healthcare in the last 3 months?
→ Sums expenses in healthcare categories and cites each receipt

What's the contact for the lawyer who helped me with the lease?
→ Searches for entity "lawyer" + context "lease" and shows phone/email

Is there an idea I noted about workplace automation?
→ Searches by meaning in idea notes and lists relevant ones

When was my last laptop maintenance?
→ Finds notes with maintenance, computer/laptop and dates

Follow-up questions

The chat is multi-turn: you can ask follow-up questions without repeating context.

You: How much did I spend on my car in 2025?
CleverNote: [lists expenses citing notes]

You: Which of those was the most expensive?
CleverNote: [identifies the largest expense from previous context]

You: Give me the phone number for that shop
CleverNote: [searches the entity linked to the cited note]

When the AI doesn’t know

If you ask about something not in your notes, the chat responds honestly that it didn’t find the information. It never invents data. The response always reflects what’s in your archive.

This is guaranteed by the architecture: the system does retrieval first, then generates the response based on what it found. Without relevant retrieval, the response explicitly indicates the absence.

See also: Semantic search, Entity-scoped chat, Multi-turn chat