> For the complete documentation index, see [llms.txt](https://verbatim-ai.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://verbatim-ai.gitbook.io/docs/stories.md). # Stories A **Story** is a conversation with your AI Corpus. You ask questions in natural language, and the Corpus answers using **only your knowledge base** โ€” then names the reference documents it relied on. Every exchange is saved so you can pick it up later. ![Just ask your AI Corpus](https://2268465679-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fo7ZuLNDm8UZdULnJ2OON%2Fuploads%2Fgit-blob-93a2df17ec57866b0b2114005281303e17943b23%2Fonboarding-ask.png?alt=media) ## The Story screen The **Story** screen centralises all your conversations, whether just started or already rich with content. Each Story appears as a card showing: * A **๐Ÿ•˜ icon**, the same for every Story, **badged with the number of Corpus** it reads from when there is more than one * The **title** of the Story * The **date and time** of the last change * The number of **posts** and **tokens** used * A menu icon (**โ‹ฎ**) for quick actions Unstarted Stories are clearly identified, and Stories in progress show their post and token counts so you stay in control of your usage. ### Quick actions โš™๏ธ From a Story's **โ‹ฎ** menu you can: * **Open and continue** the Story * **Delete** the Story from your history ## Start a Story ๐ŸŽฏ Two ways in, both landing on the same conversation: * The **โž• button** on the Story screen asks which Corpus the Story should read from. Tick one, or **tick several** โ€” the Story then answers from all of them at once. * The **Start a story** button on a [Corpus](/docs/ai-corpus.md) page skips that step and opens straight onto that Corpus. ### Ask across several Corpus ๐Ÿ”— A Story is not limited to one knowledge base. Picking your contracts *and* your invoices lets you ask questions neither could answer alone, and every answer still **names the documents it used** โ€” whichever Corpus they live in. The **Start the story** button stays inactive until at least one Corpus is ticked, and the Story takes the name and icon of the **first** one you picked. The set is fixed once the Story opens: to ask across a different one, start a new Story. > โ„น๏ธ Delete a Corpus a Story reads from and that Story becomes **read only** โ€” still readable and copyable, but no longer continued, since it would be answering from less than before. ### Set the scene ๐ŸŽฌ When you begin a Story, share the **context** of what you want to achieve โ€” your approach, your expectations, your objectives. For example: > I want to write the weekly report for the `Alpha` project. It should give an overview of the project, its schedule, its risks and its objectivesโ€ฆ > โ„น๏ธ This opening context needs a **minimum of 10 words**. Your Corpus uses it to understand the starting point of the conversation. ## What a Story reads from ๐Ÿ“š At the top of every Story, a row of chips names **each Corpus it searches**. Click one to open that Corpus. A Corpus deleted since is still listed and marked as removed โ€” the answers above it were drawn from it. The Story's ๐Ÿ•˜ icon carries the same badge in its header as on the Story screen, so a Story looks like itself wherever you meet it. ## Chat with your Corpus ๐Ÿ’ฌ From there, converse as naturally as possible. For each request, your Corpus analyses your question, searches your knowledge base and replies with an answer that **explicitly cites the documents** it used โ€” so you can trust and trace every response. ### Check the sources ๐Ÿ” Open the **sources** of an answer to see which documents it drew on, with the passage that was used and thumbnails of the pages it came from. Beside a source's name, the **๐Ÿ‘๏ธ preview button** opens that document [page by page](/docs/documents.md#previewing-a-document) โ€” the fastest way to read a citation in its context. ### Choose the agent that answers ๐Ÿค– The ๐Ÿค– button in a Story's header picks the [**agent**](/docs/agents.md) running your questions โ€” how widely your documents are searched, which excerpts survive, the instruction the model receives, and which model writes the reply. The picker lists every agent your organisation can use, with a tick on the one answering right now. Until you pick, that is the agent badged **Default**; going back to it is simply picking its card again. The button stays **lit while you are on an agent of your own**, and its tooltip always names whichever is answering. The choice is yours rather than the Story's: it holds for **this conversation and the ones after it**, and it survives closing the app. You can also set it from your [profile](/docs/profile.md), or by tapping a card in the [Agents](/docs/agents.md) list. Answers already written keep the agent that produced them. ### Choose your Corpus model Several [personalities](/docs/profile.md) are available, each with its own skills โ€” `Aly`, `Nova`, `Lexa` and more. By default a Story inherits the personality of its [Corpus](/docs/ai-corpus.md), but you can pick a different one to better match your need. ## Customise a Story โš™๏ธ Open the **Edit** view to tailor a Story: * **Name** โ€” give the Story a meaningful title (e.g. *โ€œMarketing Campaignโ€*) so you and your colleagues instantly understand its purpose. * **Objectives** ๐ŸŽฏ โ€” describe your goal so the Corpus stays focused (e.g. lead generation). * **Corpus** ๐Ÿค– โ€” choose the personality whose skills best fit the Story. > ๐Ÿ’ก Don't forget to click **Save** at the bottom of the page to apply your changes. ## Share and export ๐Ÿ“ค * **Copy** a single post โ€” or the whole Story โ€” to reuse it elsewhere or build a document from the conversation. * **Export** your Story to a **PDF** that's easy to share. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://verbatim-ai.gitbook.io/docs/stories.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.