Deep Dive Reports
Combine live web research with your internal Knowledge Base into one comprehensive, cited report — then chat with an AI Assistant over your research history.
What Is Deep Dive Reports?
Deep Dive Reports takes a single plain-language prompt and researches it two ways at once: it crawls and scrapes the live web for relevant sources, and it pulls in matching sources from your own internal Knowledge Base, then synthesizes both into one structured markdown report. Every run is saved automatically to a Previous Researches history table, and an embedded AI Assistant can analyze your research history on demand.
In one sentence
Ask a question that blends the outside world with what your organization already knows — Deep Dive Reports researches both and hands back one synthesized report, kept in a searchable history.
Web Research
A backend crawling/scraping engine discovers and reads live pages relevant to your prompt.
Internal Knowledge
Attach sources from your Knowledge Base — manually, or auto-attached by relevance — to blend in what your organization already knows.
Previous Researches
Every run — prompt, source counts, and full report — is saved automatically and listed in a multi-select history table below.
AI Assistant
Chat with your research history — summarize findings, compare reports, spot under-researched runs, or get follow-up prompt suggestions.
How It Works
On Run Deep Dive, the screen first plans and crawls web sources relevant to the prompt, then combines them with any attached internal Knowledge Base sources, and finally synthesizes everything into one cited markdown report. The run — including the report content and how many web/internal sources were used — is saved automatically to the Previous Researches history, with no separate save step required.
The report itself can additionally be pinned with Save as Report, which stores a copy in My Reports alongside reports from other TotalApp screens.
Attaching Internal Knowledge
Click Internal Knowledge next to the prompt box to open the Attach Knowledge modal, where you can manually pick sources from your Knowledge Base. If relevance ranking is available (an embedding model is configured), sources can also be auto-attached based on how closely they match your prompt — shown with a sparkle badge instead of the manual library badge.
Ranking can degrade gracefully
If no embedding model is configured, relevance ranking is unavailable — internal sources are still usable, just selected without a relevance score. The report shows a banner when this happens.
Previous Researches
Below the report panel, a Previous Researches table lists every past Deep Dive run with its prompt, combined source count (web + internal), and date. Each row supports:
- Select — a checkbox per row, plus a select-all checkbox in the header, for choosing a batch of runs to hand to the AI Assistant.
- View — reloads that run's prompt and full report back into the panel above.
- Delete — permanently removes that run from history.
When one or more rows are selected, a selection counter and a Clear selection action appear above the table.
AI Assistant
A collapsible AI Assistant panel lives on the right edge of the screen (the tab). It reads either your selected research runs from the history table, or — if nothing is selected — all previous runs, and can:
- Summarize findings across the selected research runs
- Compare or cross-reference multiple reports — contradictions, overlapping themes
- Highlight which runs used the fewest sources and may need re-running with more
- Suggest follow-up research prompts based on gaps in the existing reports
Any assistant reply can be saved with Save as Report, which stores it in My Reports tagged with the source "Deep Dive Reports Assistant".
Follows your Writer Engine setting
The AI Assistant respects whichever Writer Engine is selected in Settings → Agentic — Local CLI, Cloud API, Ollama, or an in-browser model — so it behaves consistently with every other AI feature in TotalApp.
Reports vs. Research History — What's the Difference?
Previous Researches (the history table on this screen) automatically keeps every run you've ever executed, so nothing is lost even if you never click Save. Save as Report (on the report panel, or from the AI Assistant) is a separate, explicit action that copies a specific report into TotalApp's shared report system, visible in My Reports and shareable with teammates.
Personal by default, shareable on demand
A report saved from this screen is visible only to you until you explicitly share it from My Reports — either with your whole workspace or with individually chosen people.