# How to connect Box to Grok Bot ```json { "title": "How to connect Box to Grok Bot", "toolkit": "Box MCP", "toolkit_slug": "box_mcp", "framework": "Grok Bot", "framework_slug": "grok-bot", "url": "https://composio.dev/toolkits/box_mcp/framework/grok-bot", "markdown_url": "https://composio.dev/toolkits/box_mcp/framework/grok-bot.md", "updated_at": "2026-11-12T06:01:27.448Z" } ``` ## Introduction Grok is xAI's assistant, and its bot builder lets you create a personal agent that holds context, runs skills, and connects to the apps you already use through plugins. This guide explains how to connect Box to a Grok bot using Composio Connect, which handles OAuth, token refresh, and reliability so you do not have to. You add Composio as a plugin, authenticate once, then ask your bot to work with Box in plain language. ## Also integrate Box MCP with - [ChatGPT](https://composio.dev/toolkits/box_mcp/framework/chatgpt) - [Claude Cowork](https://composio.dev/toolkits/box_mcp/framework/claude-cowork) - [Hermes](https://composio.dev/toolkits/box_mcp/framework/hermes-agent) - [Atomic Agent](https://composio.dev/toolkits/box_mcp/framework/atomic-agent) ## TL;DR Composio exposes 1,700+ apps to your Grok bot through a single connection, so you get: - Connect to 1,700+ apps with on-demand tool loading, so you do not fill your context window with unnecessary tool definitions. - Use programmatic tool calling through a remote workbench, letting the model write its own code to handle complex tool chaining. This reduces back-and-forth for frequent tool calls. - Handle large tool responses outside the model context to keep conversations lean. ## Connect Box MCP to Grok Bot ### Connecting Box to Grok Bot Once Grok Bot is set up, you connect Box entirely from its interface, and the whole flow takes under a minute. ### Set up Grok Bot Get access. Grok Bot is included with Cursor Ultra, Cursor Premium Teams, and SuperGrok Heavy. Get the desktop app for macOS or Windows from [x.ai/bot](https://x.ai/bot) and sign in. An iPhone companion app is also available, and your bots sync across devices. Create a bot. Pick a suggested role or define your own, then give it a short name, one primary job, and a short description of how it should work. Focused bots work best, because a clear role gives the bot more useful context over time. ### Connect Box ### 1. Add the Composio plugin Click the button. The Composio listing on x.ai opens and launches Grok Bot for you. If the app does not open on its own, click Add to Grok Bot. [Add to Grok Bot](https://x.ai/bot/plugin/32661537) Grok Bot opens the Composio plugin, showing its connector and skills. Click Add. ### 2. Authenticate Go to the Installed section and click the Composio entry. You'll see its connector and skills listed. Click Authenticate. You'll be taken to the Composio consent screen. Review the access it requests and click Authorize. Composio handles the OAuth handshake for you, so you never touch API keys or tokens. ### 3. Connect your Box account Back in the chat, ask your bot to connect to Box or give it any Box task. For example, ask it to: - "Search recent confidential contracts in Box" - "Download latest design files from Box" - "List shared folders for marketing team" It will prompt you to authenticate and authorize access. That's it. Composio's tools are now available in your Grok bot, and your Box account is ready to use. ## What is Grok Bot? Grok is xAI's assistant, available as a chat interface and a bot builder for creating personal agents that run skills and connect to external apps. With Composio Connect, your Grok bot can securely access apps like Box through MCP without you sharing account credentials directly with the agent. ## What is the Box MCP server, and what's possible with it? The Box MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Box account. It provides structured and secure access to your enterprise files, folders, comments, tasks, collaborations, and metadata, so your agent can search, retrieve, analyze, organize, and upload content on your behalf. - Content search and retrieval: Instruct your agent to find files by keywords or metadata, search for folders by name, and retrieve file content or details. - File and folder organization: Have your agent create folders, copy files or folders, list folder contents, and upload new files to Box. - Document review and analysis: Let the agent display file previews, inspect the current preview page, and answer follow-up questions about visible text, figures, or other content. - Comments, tasks, and collaboration tracking: Direct your agent to add or list file comments, review tasks linked to a file, and identify collaborators on files and folders. - Enterprise metadata management: Have your agent list metadata templates, inspect their schemas, create new templates, and find files using metadata queries. ## Supported Tools | Tool slug | Name | Description | |---|---|---| | `BOX_MCP_COPY_FILE` | Copy file | Copies an existing file to a destination folder in Box (see Box API Copy file). The source file is not modified. Provide `parent_folder_id` as the folder that will contain the new copy; use `"0"` for the user's root folder. Optionally set `name` to rename the copy; otherwise the original file name is used. **Key features:** - Optional new name for the copied file Copy a file into a folder: { "file_id": "123456789", "parent_folder_id": "987654321" } Copy a file to the user's root folder with a new name: { "file_id": "123456789", "parent_folder_id": "0", "name": "Quarterly-report-copy.docx" } | | `BOX_MCP_COPY_FOLDER` | Copy folder | Copies an existing folder to a destination parent folder in Box (see Box API Copy folder). The source folder is not modified. Provide `parent_folder_id` as the folder that will contain the new copy; use `"0"` for the user's root folder. The root folder (`folder_id` `"0"`) cannot be copied. Optionally set `name` for the new copy; otherwise the original folder name is used. **Key features:** - Optional new name for the copied folder Copy a folder into another folder: { "folder_id": "123456789", "parent_folder_id": "987654321" } Copy a folder to the user's root with a new name: { "folder_id": "123456789", "parent_folder_id": "0", "name": "Project-backup" } | | `BOX_MCP_CREATE_FILE_COMMENT` | Create file comment | Creates a new comment on a specific file. This tool allows you to add comments to files for collaboration, feedback, or documentation purposes. Comments are visible to all users with access to the file and can be used for team communication and file annotation. The message can include special characters, line breaks, and Unicode characters. The tool returns complete comment information including metadata about the comment creation, author, and timestamps. Basic file comment: { "file_id": "123456789", "message": "This document looks good overall. Please review the financial projections in section 3." } Multi-line structured comment: { "file_id": "777888999", "message": "Review Summary:\\n\\nContent: Excellent\\nStructure: Well organized\\nFormatting: Minor issues\\nCitations: Missing sources\\n\\nOverall: Good work, please address the citations before final submission." } | | `BOX_MCP_CREATE_FOLDER` | Create folder | Creates new folders in Box. Create folder in root: { "name": "Project Documents" } Create folder in specific parent: { "name": "Q4 Reports", "parent_folder_id": "123456789" } | | `BOX_MCP_CREATE_METADATA_TEMPLATE` | Create metadata template | Creates a new enterprise metadata template. scope must be "enterprise"; each field needs type, key, and display_name. Optional template_key, hidden, copy_instance_on_item_copy, enum/multiSelect/taxonomy. Enum color_id optional (0=Yellow, 1=Orange, 2=Red, 3=Purple, 4=LightBlue, 5=DarkBlue, 6=Green, 7=Grey); omit for default Yellow (0). SDK create sends keys only; non-default colors are PATCHed after create when color_id is set. On success returns created template JSON (or JSON after color PATCH). If color PATCH fails, response still succeeds with colorApplyWarning. Minimal template with one string field: { "scope": "enterprise", "display_name": "Sample contract template", "template_key": "sampleContract", "fields": [ { "type": "string", "key": "Name", "display_name": "Name" } ] } { "id": "f7a9891f", "scope": "enterprise_12345", "templateKey": "sampleContract", "displayName": "Sample contract template", "hidden": false, "fields": [ { "type": "string", "key": "Name", "displayName": "Name" } ] } Enum field with classification colors (color_id optional per option): { "scope": "enterprise", "display_name": "Data classes", "template_key": "dataClasses", "fields": [ { "type": "enum", "key": "tier", "display_name": "Tier", "options": [ { "key": "Public", "color_id": 6 }, { "key": "Internal", "color_id": 4 }, { "key": "Restricted", "color_id": 2 } ] } ] } | | `BOX_MCP_GET_FILE_CONTENT` | Get file content | Get the text content of a file from Box. Prefers the markdown representation (preserves document structure for better LLM use) and falls back to extracted text when markdown is not available. If neither representation is available or both are still pending, an error is returned. Using this tool is preferred over the ai_qa_single_file tool since it is faster and more efficient. Works for files with text representations such as PDFs, documents, spreadsheets, presentations, code, etc. Optional version_id reads a past version (GET /files/{file_id}/versions/{version_id}); omit it for the current version. Use list_file_versions to get version_id values. **Key Features:** - Markdown representation preferred (structure preserved for LLMs); fallback to extracted text - Optional version_id for a past file version - 50MB file size limit enforcement - Clear error messages for unsupported operations Extract text content from a file: { "file_id": "123456789", } Extract text content from a past file version: { "file_id": "123456789", "version_id": "987654321", } { "content": [ { "type": "text", "text": "This is the extracted text content from the file..." } ] } | | `BOX_MCP_GET_FILE_DETAILS` | Get file details | Gets comprehensive file information from Box. Returns essential fields by default (id, name, extension, createdBy, modifiedBy, createdAt, modifiedAt). Use 'fields' parameter to request additional properties. **Common useful fields:** size, description, tags, metadata, permissions, shared_link, parent, path_collection, file_version, sha1, lock, classification, comment_count, has_collaborations, representations, watermark_info, uploader_display_name, content_created_at, content_modified_at, item_status, expires_at, trashed_at. Basic file info: {"file_id": "123456789"} File with metadata and permissions: { "file_id": "987654321", "fields": ["id", "name", "size", "permissions", "tags", "metadata", "shared_link"] } Detailed file analysis: { "file_id": "555666777", "fields": ["id", "name", "description", "file_version", "sha1", "path_collection", "classification", "comment_count", "has_collaborations"] } | | `BOX_MCP_GET_FILE_PREVIEW` | Get file preview | Displays an interactive preview widget for a Box file directly in chat. Supports: ai, bmp, cr2, crw, csv, dng, doc, docx, dwg, eml, eps, gdoc, gif, gsheet, gslide, gslides, heic, htm, html, idml, indd, indt, inx, jpeg, jpg, key, nef, numbers, odp, ods, odt, pages, pdf, png, ppt, pptx, ps, psd, raf, raw, rtf, svg, svs, tga, tif, tiff, tsv, webp, wpd, xhtml, xls, xlsm, xlsx. PDF uses direct download; other types use the appropriate representation (e.g. PDF for documents, JPG/PNG for images). Use when user wants to view file contents without downloading. For file metadata, use get_file_info instead. If the preview payload would exceed 3MB (full file for PDFs; generated representation for other types), do not use this. Do not attempt to generate an interactive artifact. When a preview is open, the widget sends file preview context (fileId, fileName, currentPage) into the conversation. If the user asks about "this graph", "explain this", "what's on this page", or similar, look for that context and call get_preview_page(fileId, currentPage) to fetch the page image—do not ask the user which page they are on. | | `BOX_MCP_GET_FOLDER_DETAILS` | Get folder details | Retrieves comprehensive information about a folder in Box. Returns essential fields by default (id, name, createdBy, modifiedBy, createdAt, modifiedAt). Use 'fields' parameter to request additional properties. **Common useful fields:** size, description, tags, metadata, permissions, shared_link, parent, path_collection, folder_upload_email, has_collaborations, allowed_invitee_roles, can_non_owners_invite, can_non_owners_view_collaborators, is_collaboration_restricted_to_enterprise, classification, allowed_shared_link_access_levels, content_created_at, content_modified_at, item_status, sync_state, watermark_info, trashed_at. **Note:** Only request 'item_collection' when the user explicitly asks for folder contents. Use 'list_folder_content_by_folder_id' for browsing folder items instead. Basic folder info: {"folder_id": "123456789"} Folder with permissions and collaboration settings: { "folder_id": "987654321", "fields": ["id", "name", "size", "permissions", "can_non_owners_invite", "has_collaborations"] } Detailed folder analysis with settings: { "folder_id": "555666777", "fields": ["id", "name", "description", "path_collection", "folder_upload_email", "allowed_invitee_roles", "classification", "shared_link", "tags"] } | | `BOX_MCP_GET_METADATA_TEMPLATE_SCHEMA` | Get metadata template schema | Returns the schema for a metadata template. Shield template "securityClassification-6VMVochwUWo": option colorID is normalized from staticConfig (default Yellow=0). Palette: 0=Yellow, 1=Orange, 2=Red, 3=Purple, 4=LightBlue, 5=DarkBlue, 6=Green, 7=Grey. | | `BOX_MCP_GET_PREVIEW_PAGE` | Get preview page | Use when user asks follow-up question to a file previewed with get_file_preview ("Explain this...", "What does ____ mean?")Returns the current preview page as an image so you can analyze it (figures, text, charts, etc.). You cannot see the page content until you call this tool. Required workflow: 1. Find the file preview context in the conversation. The preview widget sends it as content like: "Current file preview: fileId=123, fileName=doc.pdf, currentPage=3. Use get_preview_page with this fileId and currentPage...". Look for this before calling. 2. Extract fileId and currentPage from that context. If the user asks about "this page", "these figures", "what's on the page", use currentPage from the context. If they name a page (e.g. "page 7"), use that number instead. 3. Call get_preview_page(fileId, currentPage) to fetch the page image. You must call this tool to see the page—there is no other way to view its content. 4. Use the returned image to answer the user's question. Do not call with a fileId or page from elsewhere. If no file preview context is present (no "Current file preview: fileId=..." in the conversation), do not call. Supported extensions: ai, bmp, cr2, crw, csv, dng, doc, docx, dwg, eml, eps, gdoc, gif, gsheet, gslide, gslides, heic, htm, html, idml, indd, indt, inx, jpeg, jpg, key, nef, numbers, odp, ods, odt, pages, pdf, png, ppt, pptx, ps, psd, raf, raw, rtf, svg, svs, tga, tif, tiff, tsv, webp, wpd, xhtml, xls, xlsm, xlsx. | | `BOX_MCP_LIST_FILE_COMMENTS` | List file comments | Lists all comments on a specific file. Returns a paginated list of comments with configurable fields. Default fields (id, type, message, created_by, created_at) are always returned, but you can specify additional fields as needed. Supported fields include: id, type, message, created_at, created_by, modified_at, tagged_message, and any other comment-related fields supported by the Box API. Use pagination parameters (limit/offset) to handle files with many comments efficiently. Basic file comments listing: { "file_id": "123456789", "limit": 50 } List with specific fields: { "file_id": "987654321", "fields": ["id", "message", "created_at", "created_by", "modified_at"], "limit": 100 } Paginated comments listing: { "file_id": "555666777", "fields": ["id", "type", "message", "tagged_message", "created_by"], "limit": 25, "offset": 50 } | | `BOX_MCP_LIST_FOLDER_CONTENT_BY_FOLDER_ID` | List folder content by folder | Lists files, folders, and web links contained in a folder. Returns a paginated list of items with at least fields. Fields (id, type, name, etag) are always returned. Use folder_id "0" for the root folder. Supported fields include: id, type, name, etag, sequence_id, sha1, content_created_at, content_modified_at, created_at, created_by, modified_at, modified_by, owned_by, parent, path_collection, size, trashed_at, item_status, description, shared_link, classification, comment_count, disposition_at, expires_at, extension, has_collaborations, is_accessible_via_shared_link, is_associated_with_app_item, is_externally_owned, is_package, lock, metadata, permissions, representations, shared_link_permission_options, tags, uploader_display_name, version_number, watermark_info, file_version. Basic folder listing: { "folder_id": "0", "limit": 50 } List with specific fields: { "folder_id": "123456789", "fields": ["id", "name", "type", "size", "modified_at", "created_by"], "limit": 100 } Paginated listing: { "folder_id": "987654321", "fields": ["id", "name", "type", "path_collection"], "limit": 30, "offset": 60 } | | `BOX_MCP_LIST_ITEM_COLLABORATIONS` | List item collaborations | Lists all collaborations for multiple files and folders in a single request. This tool efficiently aggregates collaboration information across different item types, making multiple API calls as needed and providing comprehensive results with error handling for items that cannot be accessed. The tool processes up to 10 items simultaneously, handling both files and folders. It returns detailed collaboration information including user details, roles, status, and timestamps. Partial failures are handled gracefully - if some items fail to load, the tool still returns collaborations for successful items along with error details for failed requests. Basic file collaboration listing: { "items": [ { "id": "123456789", "type": "file" } ] } Single folder collaboration listing: { "items": [ { "id": "987654321", "type": "folder" } ] } Multiple items for bulk analysis: { "items": [ { "id": "111222333", "type": "file" }, { "id": "444555666", "type": "file" }, { "id": "777888999", "type": "folder" }, { "id": "101112131", "type": "folder" } ] } | | `BOX_MCP_LIST_METADATA_TEMPLATES` | List metadata templates | Lists available Box metadata templates so you can discover scope and templateKey for search_files_metadata. Use this first to find templates, then get_metadata_template_schema to get field keys. Call with scope "enterprise" for your org's templates or "global" for Box global templates. | | `BOX_MCP_LIST_TASKS` | List tasks | Lists all tasks associated with a specific file. This tool retrieves task information including task details, message, status, and completion rule, due date from the Box platform. Tasks in Box are used to assign work to users on files, allowing for collaboration and workflow management. This tool provides access to all tasks assigned to a particular file, helping with project management and task tracking. The response includes comprehensive task details such as task type, assignee information, due dates, completion status, and any associated comments or messages. List tasks for a file: { "file_id": "1234567890" } | | `BOX_MCP_SEARCH_FILES_KEYWORD` | Search files keyword | Searches for files using keywords and supports metadata filters, file extension filtering (without dots), and field selection. Maps to Box's searchForContent API with comprehensive filtering capabilities. The response always includes core fields: id, name, description, type, and size. Each entry also includes `url` (Box web link for files and folders; web links use the API `url` when returned). The 'fields' parameter allows you to request additional fields beyond these core ones. All requested fields (core + additional) will be included in the API request and returned in the response. **Available Additional Fields:** fileVersion, sha1, contentCreatedAt, contentModifiedAt, createdAt, createdBy, modifiedAt, modifiedBy, ownedBy, parent, pathCollection, purgedAt, sharedLink, size, trashedAt, classification, commentCount, dispositionAt, expiresAt, extension, hasCollaborations, isAccessibleViaSharedLink, isAssociatedWithAppItem, isExternallyOwned, isPackage, lock, metadata, permissions, representations, sharedLinkPermissionOptions, tags, uploaderDisplayName, watermarkInfo **Date Range Filters:** Date range filters properties of the input schema: created_at_range, updated_at_range, deleted_at_range. Use comma-separated RFC3339 timestamps. **Format:** "start_date,end_date" where: - Omit start date to search all items before end date: ",2023-12-31T23:59:59Z" - Omit end date to search from start date to the present: "2023-01-01T00:00:00Z," - Both dates for specific range: "2023-01-01T00:00:00Z,2023-12-31T23:59:59Z" Basic keyword search (returns core fields only): { "query": "contract agreement", "limit": 50 } Advanced search with additional fields: { "query": "quarterly report", "file_extensions": ["pdf", "docx"], "ancestor_folder_id": "123456789", "fields": ["createdAt", "modifiedAt", "size", "createdBy"] } Searching by date: { "query": "quarterly report", "file_extensions": ["pdf", "docx"], "ancestor_folder_id": "123456789", "created_at_range”: "2023-01-01T00:00:00Z,2023-12-31T23:59:59Z", "fields": ["createdAt"] } Search with metadata filters and additional fields: { "query": "invoice", "file_extensions": ["pdf"], "limit": 30, "fields": ["contentCreatedAt", "pathCollection", "metadata"], "mdfilters": [ { "scope": "enterprise", "templateKey": "invoiceTemplate", "filters": { "amount": { "gt": 1000, "lt": 5000 }, "status": "approved", "department": "finance" } } ] } Multi-template metadata search with custom fields: { "query": "project document", "fields": ["modifiedBy", "sharedLink", "tags", "permissions"], "mdfilters": [ { "scope": "enterprise", "templateKey": "projectTemplate", "filters": { "priority": "high", "deadline": { "gt": "2024-01-01" } } }, { "scope": "global", "templateKey": "reviewTemplate", "filters": { "reviewer": "john.doe@company.com" } } ] } | | `BOX_MCP_SEARCH_FILES_METADATA` | Search files metadata | Searches for files using SQL-like metadata queries. Requires: from (e.g. enterprise_123456.templateKey from list_metadata_templates), fields (field keys from get_metadata_template_schema). ancestor_folder_id defaults to "0" (root) if omitted. Use list_metadata_templates and get_metadata_template_schema first to discover templates and field keys. | | `BOX_MCP_SEARCH_FOLDERS_BY_NAME` | Search folders by name | Searches for folders by name within Box using keyword matching. This tool helps locate specific folders by their names and can be scoped to search within a particular parent folder. Returns basic folder information including ID, type, and name for each matching folder. The search performs a keyword-based match against folder names and can be limited to search within a specific ancestor folder. This is particularly useful for navigating complex folder structures, finding project folders, or locating organizational folders by name patterns. The tool uses Box's search API with folder type filtering, ensuring only folders (not files) are returned in results. Search results are limited to preserve performance and can be paginated using the limit parameter. **Date Range Filters:** Date range filters properties of the input schema: created_at_range, updated_at_range, deleted_at_range. Use comma-separated RFC3339 timestamps. **Format:** "start_date,end_date" where: - Omit start date to search all folders before end date: ",2023-12-31T23:59:59Z" - Omit end date to search from start date to the present: "2023-01-01T00:00:00Z," - Both dates for specific range: "2023-01-01T00:00:00Z,2023-12-31T23:59:59Z" Basic folder search: { "folder_name": "Project Documents", "limit": 30 } Search within specific parent folder: { "folder_name": "2024 Reports", "ancestor_folder_id": "123456789", "limit": 50 } Search for folders with partial name match: { "folder_name": "Finance", "limit": 20 } Limited results search: { "folder_name": "Marketing Campaign", "ancestor_folder_id": "987654321", "limit": 10 } Search in root directory and limit results by date: { "folder_name": "Shared Resources", "created_at_range": "2023-01-01T00:00:00Z,2023-12-31T23:59:59Z" } | | `BOX_MCP_UPLOAD_FILE` | Upload file | Uploads a new file to Box. This tool creates a new file in Box using the provided text content. The file name including the extension must be provided and the extension must match the file content. Markdown uploads (e.g. .md) are regular text files like any other supported extension; use a .boxnote file name only when file_content should become a native Box Note via the notes conversion API. A parent folder ID can optionally be provided to determine where the new file should be created. The metadata details of the newly uploaded file are returned in response. If a file already exists with the same name, the tool will return an error. **Key Features:** - File creation with text content - Supported file extensions: txt, md, boxnote, html, svg, xml, csv, json, js, ts, py, sh - If the file extension is invalid, a .txt file will be created **Important:** This tool can only accept text content, to upload binary files, use the `get_upload_url` tool instead. Upload a file to Box: { "file_name": "example.md", "file_content": "#This is my example markdown file", "parent_folder_id": "0", } { "content": [ { "type": "text", "text": "{"file_link": "https://example.app.box.com/file/123456789", "file_id": "123456789", "file_name": "example.md", "size": 100, "created_at": "2025-01-01T00:00:00Z"}" } ] } | | `BOX_MCP_WHO_AM_I` | Who am i | Returns detailed information about the currently authenticated Box user. This tool provides comprehensive user profile data including basic identification, contact information, role details, and account settings. No input parameters are required as it retrieves information for the authenticated user making the request. This tool is essential for understanding the current user context, verifying authentication status, and obtaining user-specific information for personalization or access control purposes. It's particularly useful when building user-aware applications or when you need to identify the current user's permissions and organizational role. The response includes core user attributes such as ID, name, email, login, role, job title, and additional account details like space usage, language preferences, and timezone settings. Get current user information: {} | ## Supported Triggers None listed. ## Creating MCP Server - Stand-alone vs Composio SDK Once connected, your Grok bot can access the Box MCP server through Composio to run the actions you authorize, using plain language in the chat. ## Complete Code None listed. ## Conclusion ### Way Forward Now that Box is connected, extend your setup by connecting the other apps you already use every day, so your Grok bot can run true cross-app workflows end to end. - Connect Calendar to turn threads into scheduled meetings automatically. - Connect Slack or Teams to post summaries, approvals, and alerts where your team works. - Connect Notion, Linear, Jira, or Asana to convert requests into tickets, tasks, and docs. - Connect Drive, Dropbox, or OneDrive to fetch, file, and share attachments without manual steps. Start with one workflow you do repeatedly, then keep adding apps as you find new handoffs. With everything behind a single connection, your bot can coordinate multiple tools safely and reliably in one conversation. ## How to build Box MCP Agent with another framework - [ChatGPT](https://composio.dev/toolkits/box_mcp/framework/chatgpt) - [Claude Cowork](https://composio.dev/toolkits/box_mcp/framework/claude-cowork) - [Hermes](https://composio.dev/toolkits/box_mcp/framework/hermes-agent) - [Atomic Agent](https://composio.dev/toolkits/box_mcp/framework/atomic-agent) ## Related Toolkits - [Google Drive](https://composio.dev/toolkits/googledrive) - Google Drive is a cloud storage platform for uploading, sharing, and collaborating on files. It's perfect for keeping your documents accessible and organized across devices. - [Google Docs](https://composio.dev/toolkits/googledocs) - Google Docs is a cloud-based word processor that enables document creation and real-time collaboration. Its seamless sharing and version history make team editing and content management a breeze. - [Google Super](https://composio.dev/toolkits/googlesuper) - Google Super is an all-in-one suite combining Gmail, Drive, Calendar, Sheets, Analytics, and more. It gives you a unified platform to manage your digital life, boosting productivity and organization. - [ActiveMerge](https://composio.dev/toolkits/active_merge) - ActiveMerge is a document automation platform for generating documents, images, merged files, and e-signature workflows. It helps teams turn templates and stored files into finished, shareable documents faster. - [Affinda](https://composio.dev/toolkits/affinda) - Affinda is an AI-powered document processing platform that automates data extraction from resumes, invoices, and more. It streamlines document-heavy workflows by turning files into structured, actionable data. - [Agility cms](https://composio.dev/toolkits/agility_cms) - Agility CMS is a headless content management system for building and managing digital experiences across platforms. It lets teams update content quickly and deliver omnichannel experiences with ease. - [Airparser](https://composio.dev/toolkits/airparser) - Airparser is an AI-powered parser for extracting structured data from emails, documents, PDFs, and files. It turns messy inbound content into clean fields your apps and workflows can use. - [Algodocs](https://composio.dev/toolkits/algodocs) - Algodocs is an AI-powered platform that automates data extraction from business documents. It delivers fast, secure, and accurate processing without templates or manual training. - [Api2pdf](https://composio.dev/toolkits/api2pdf) - Api2Pdf is a REST API for generating PDFs from HTML, URLs, and documents using powerful engines like wkhtmltopdf and Headless Chrome. It streamlines document conversion and automation for developers and businesses. - [Archive.org](https://composio.dev/toolkits/archive_org) - Archive.org is the Internet Archive's digital library of websites, books, audio, video, and images. Use it to preserve, explore, and access historical web captures and public-domain media. - [Aryn](https://composio.dev/toolkits/aryn) - Aryn is an AI-powered platform for parsing, extracting, and analyzing data from unstructured documents. Use it to automate document processing and unlock actionable insights from your files. - [Boldsign](https://composio.dev/toolkits/boldsign) - Boldsign is a digital eSignature platform for sending, signing, and tracking documents online. Organizations use it to automate agreements and manage legally binding workflows efficiently. - [Boloforms](https://composio.dev/toolkits/boloforms) - BoloForms is an eSignature platform built for small businesses, offering unlimited signatures, templates, and forms. It simplifies digital document signing and team collaboration at a predictable, fixed price. - [Carbone](https://composio.dev/toolkits/carbone) - Carbone is a blazing-fast report generator that turns JSON data into PDFs, Word docs, spreadsheets, and more using flexible templates. It lets you automate document creation at scale with minimal code. - [Castingwords](https://composio.dev/toolkits/castingwords) - CastingWords is a transcription service specializing in human-powered, accurate transcripts via a simple API. Get seamless audio-to-text conversion for interviews, meetings, podcasts, and more. - [CertSeal](https://composio.dev/toolkits/cert_seal) - CertSeal is an API platform for issuing and managing verifiable digital certificates. Use it to automate trusted certificate workflows for recipients, batches, designs, and webhooks. - [Cloudconvert](https://composio.dev/toolkits/cloudconvert) - CloudConvert is a powerful file conversion service supporting over 200 file formats. It streamlines converting, compressing, and managing documents, media, and more, all in one place. - [Cloudlayer](https://composio.dev/toolkits/cloudlayer) - Cloudlayer is a document and asset generation service for creating PDFs and images via API or SDKs. It lets you automate high-quality doc creation, saving dev time and reducing manual work. - [CloudPDF](https://composio.dev/toolkits/cloud_pdf) - CloudPDF is an API platform for creating, uploading, managing, and sharing PDF documents. Use it to build secure PDF workflows with webhooks, share links, and access control. - [Cloudpress](https://composio.dev/toolkits/cloudpress) - Cloudpress is a content export tool for Google Docs and Notion. It automates publishing to your favorite Content Management Systems. ## Frequently Asked Questions ### What are the differences in Tool Router MCP and Box MCP? With a standalone Box MCP server, the agents and LLMs can only access a fixed set of Box tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Box and many other apps based on the task at hand, all through a single MCP endpoint. ### Can I use Tool Router MCP with Grok Bot? Yes, you can. Grok Bot fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Box tools. ### Can I manage the permissions and scopes for Box while using Tool Router? Yes, absolutely. You can configure which Box scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do. ### How safe is my data with Composio Tool Router? All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Box data and credentials are handled as safely as possible. --- [See all toolkits](https://composio.dev/toolkits) · [Composio docs](https://docs.composio.dev/llms.txt)