> ## Documentation Index
> Fetch the complete documentation index at: https://miniti.app/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# investigations

> research a live meeting question with openai, cited web sources, or selected mac code

investigations let you research something raised during a live meeting without leaving miniti. nothing runs automatically - every openai request starts with your click or tap.

## how suggestions work

miniti checks the latest finalized transcript text locally for phrases such as "could we build this?", "look into why this fails", or "is there a way to automate this?". when it finds one, it can show a dismissible **worth investigating** prompt.

this check is a small on-device phrase rule. it does not call an ai model, search the web, or use api quota. you can also use **investigate** at any time without waiting for a suggestion.

## research the web

available on macos and ios:

1. start a meeting and wait for some transcript text
2. choose **investigate**
3. on macos, choose **research the web**. on ios, the button starts web research directly
4. keep talking while openai researches the recent question
5. review the answer and open any cited source links inside the result sheet

use **run again** to refresh the result. closing the sheet cancels the active request, and a result from an older meeting cannot replace the current one.

## investigate a codebase on macos

1. open **settings → ai & models → meeting investigations**
2. choose one codebase folder
3. during a meeting, choose **investigate → investigate codebase**

miniti searches the selected folder locally and sends only bounded excerpts from relevant files to openai. it does not grant openai filesystem access. the result lists the files reviewed.

the local scan:

* skips hidden, generated, dependency, build, and symlinked content
* reads no more than 1,200 supported source or documentation files, up to 300 kb each
* sends excerpts from no more than 8 ranked files, capped at 40,000 characters total
* redacts likely credential assignments and private-key blocks before sending

redaction is a safety layer, not a guarantee. select only a folder whose relevant source content you are comfortable sending to openai. clear the folder from the same settings section to revoke future access.

## what is sent

an investigation can include:

* the question or statement you chose
* meeting title
* current meeting notes, including locally stored calendar prep notes carried into the meeting
* up to 14 recent finalized transcript turns
* for a mac codebase investigation, the bounded local excerpts described above

managed mode sends this request through miniti's api to openai. byok sends it directly from your device to your openai account. web research enables openai's web search tool; codebase research does not.

investigations use pinned `gpt-5.4-mini-2026-03-17`. they do not consume transcription minutes, but byok users pay openai for their own requests.

## related

* [google calendar](/docs/integrations/google-calendar) - add prep notes before a meeting
* [ai insights](/docs/features/ai-insights) - the normal live and saved insight views
* [byok](/docs/pricing/byok) - use your own openai account