> For the complete documentation index, see [llms.txt](https://planck-ai.gitbook.io/planck-ai-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://planck-ai.gitbook.io/planck-ai-docs/documentation/guides/chat-with-documents.md). # Chat with Documents Use document chat when you want to ask questions about files inside a workspace. Planck AI can work with uploaded documents as well as files synced from connected systems such as Microsoft 365, OneDrive, and SharePoint. ### Supported document types Depending on your workspace setup, Planck AI can work with files such as: * PDF files * Excel files * CSV files * Text files * Synced SharePoint files * Synced OneDrive files Support may vary by deployment and enabled integrations. ### How documents are added Documents can be added to a workspace in two main ways. #### Uploading files Users can upload documents directly into a workspace. This is useful for: * Reports * Invoices * Contracts * Spreadsheets * Project files * Operational documents #### Syncing cloud folders Admins can connect cloud storage systems such as Microsoft 365, OneDrive, or SharePoint. Once connected, selected folders can be synced into a workspace. Users can then ask questions about the synced files, subject to their permissions. ### Asking questions about a document When a document is available in a workspace, open the workspace chat and ask a clear question. Examples: ``` Summarize this document. ``` ``` What are the key risks mentioned in this contract? ``` ``` Which invoices are above 10,000 euros? ``` ``` What are the main changes compared to the previous version? ``` ``` Find documents that mention delayed rollout milestones. ``` ### Asking questions across multiple documents You can also ask questions across more than one document, depending on the files available in the workspace. Examples: ``` Compare these two spreadsheets. ``` ``` Which contracts mention automatic renewal? ``` ``` Find all documents that refer to supplier penalties. ``` ``` Summarize the key issues across all rollout reports. ``` ### Working with spreadsheets and CSV files For Excel and CSV files, Planck AI can help users analyze structured data. Examples: ``` Show the largest suppliers by spend. ``` ``` Group this data by month. ``` ``` Which rows have missing values? ``` ``` Calculate total invoice value by customer. ``` ``` Show this as a table. ``` ### Understanding answers Depending on the document and question, Planck AI may return: * A summary * Extracted values * A table * A comparison * A chart * Source references * Follow-up suggestions For important work, always review the supporting context and source material before taking action. ### Source references When Planck AI answers from documents, it may include source references or supporting context. Use these references to check where the answer came from. This is especially important for: * Contracts * Financial documents * Compliance documents * Customer documents * Operational reports * Invoices ### Permissions Users can only access documents they are permitted to see. For cloud-synced documents, Planck AI can enforce source-level access controls. This means users only see results for files they are allowed to access in the original cloud system. For example, if a user does not have access to a SharePoint document, that document should not appear in their search results. ### Best practices Use specific questions. Good examples: ``` Summarize the payment terms in this contract. ``` ``` List invoices above 10,000 euros and group them by supplier. ``` ``` Find rollout reports that mention missing permits. ``` Less useful examples: ``` Analyze this. ``` ``` What is in here? ``` ``` Tell me everything. ``` Use clear file names where possible. Good examples: ``` Supplier_Invoice_Summary_March_2026.xlsx ``` ``` Network_Rollout_Status_Q1_2026.pdf ``` Avoid vague file names such as: ``` final.pdf ``` ``` new_file.xlsx ``` ### Troubleshooting #### Document is not appearing in chat The document may still be processing, or it may not be linked to the workspace. #### Document answer looks incomplete Try asking a more specific question or checking whether the relevant file is available in the workspace. #### User cannot see synced documents For cloud-synced files, the user may need to link their cloud account or may not have access to the file in the original source system. #### Spreadsheet answer seems wrong Check whether the spreadsheet has clear headers, consistent values, and the expected sheet or tab structure. #### Files size or file type not supported Right now there is a limit of 100Mb per file that one can upload for analysis. We currently only support the following file type (pdf, doc, xlsx, txt). We do not support image files --- # 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: ``` GET https://planck-ai.gitbook.io/planck-ai-docs/documentation/guides/chat-with-documents.md?ask= ``` The question should be specific, self-contained, and written in natural language. 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.