Manage Basecamp content from an MCP client: authenticate, browse and edit projects, messages, todos, comments, kanban cards, docs/files, check-ins, and activity.
Authenticate via OAuth (basecamp_login), logout, and check current user (basecamp_whoami, basecamp_get_me).
List/get projects, people, and message types.
Messages: list, get, create, update (full or partial HTML content with append/prepend/search-replace).
TODOs: get todo set, list todos, create/update/complete/uncomplete todos with due dates, assignees, and rich content.
Comments: list, create, and update comments on any resource.
Kanban: list columns/cards, get/create/update cards with checklist steps, move cards.
Activity: cross-project recording search and campfire message browsing with filters (type, person, date, text, project, status).
Docs & Files: manage vaults, documents, and uploads; download inline blobs; upload files/URLs to get embeddable sgids.
Check-ins: list questions/answers, get details, and post answers.
Supports rich text with mentions by person ID and attachments via <bc-attachment> tags.
Enables comprehensive project management through Basecamp's API, including managing projects, messages, todos, comments, people, and kanban boards with support for creating, reading, updating content and handling pagination.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Basecamp MCP Server list active todos in the 'Website Redesign' project"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Model Context Protocol (MCP) server for Basecamp. Gives LLMs tools for projects, messages, todos, comments, people, kanban boards, docs & files, check-ins, and campfire chat.
48 tools, published on npm and installable with one npx command: no cloning, no virtualenv, no manual OAuth script to run.
Zero-install setup: npx basecamp-mcp@latest runs the server directly from npm. Authentication is one MCP tool call (basecamp_login) that opens a browser; there's no separate script to clone and run by hand.
Full Docs & Files support: read and write vaults (folders), documents, and uploads, and download inline <bc-attachment> blobs embedded in rich text. Images come back inline, text files as text, everything else saved to disk. Upload a local file or a remote URL to get the sgid that embeds it in rich text.
Check-ins (Q&A) support: list automatic check-in questions and their answers, or post new answers programmatically.
Granular content editing: messages, comments, documents, and kanban cards all support append, prepend, and search-replace operations, not just full-text replacement, so an LLM can make a small edit without resending the whole document.
Mentions by person ID: every tool that writes rich text takes <bc-attachment person-id="123"> anywhere in the HTML and turns it into a real mention. An unknown ID is an error and nothing is posted, instead of a broken mention that notifies nobody.
Cross-project activity feed: basecamp_list_recordings searches across every project by type, person, date range, and free text in one call, with automatic response-size management and pagination.
Type-safe end to end: written in TypeScript with Zod schemas validating every tool input.
Tested against the real API: the test suite exercises every tool category (messages, todos, kanban, comments, docs/files, check-ins, campfires, activity) against a live Basecamp account, not mocks.
Related MCP server: Basecamp MCP Server
The Basecamp MCP server requires Node.js 18+ and works with various MCP clients including Claude Code CLI, Claude Desktop, Cursor, VS Code, and others.
You need a Basecamp OAuth app. Register one at 37signals Launchpad with the redirect URI set to http://localhost:7652/callback.
Add the MCP server to your client with your OAuth credentials:
{
"mcpServers": {
"basecamp": {
"command": "npx",
"args": ["-y", "basecamp-mcp@latest"],
"env": {
"BASECAMP_CLIENT_ID": "your_client_id",
"BASECAMP_CLIENT_SECRET": "your_client_secret"
}
}
}
}Claude Code CLI:
claude mcp add basecamp npx basecamp-mcp@latest \
-e BASECAMP_CLIENT_ID=your_client_id \
-e BASECAMP_CLIENT_SECRET=your_client_secretClaude Desktop: Follow the MCP install guide using the JSON config above.
Cursor: Add configuration through Settings → Tools & Integrations → New MCP Server.
VS Code:
code --add-mcp '{"name":"basecamp","command":"npx","args":["-y", "basecamp-mcp@latest"]}'Once the MCP server is running, authenticate using the built-in login tool:
Call basecamp_login to open a browser window for Basecamp authorization
Authorize the app in your browser
If you have multiple Basecamp accounts, call basecamp_login again with the desired account_id
Done! Credentials are saved to ~/.config/basecamp-mcp/credentials.json
Use basecamp_whoami to check who you're logged in as, and basecamp_logout to remove stored credentials.
The server requires these environment variables:
BASECAMP_CLIENT_ID — Your Basecamp OAuth client ID
BASECAMP_CLIENT_SECRET — Your Basecamp OAuth client secret
basecamp_login - Authenticate with Basecamp via OAuth browser flow
basecamp_logout - Remove stored credentials
basecamp_whoami - Show the currently authenticated user
basecamp_list_projects - List all accessible projects with optional filtering
basecamp_get_project - Get detailed project information including dock configuration
basecamp_list_messages - List messages in a message board with optional filtering
basecamp_list_message_types - List available message types/categories for a project
basecamp_get_message - Get single message details
basecamp_create_message - Create new message with optional category and draft status
basecamp_update_message - Update message with advanced content editing (supports full replacement, append, prepend, search/replace)
basecamp_get_todoset - Get todo set container with all todo lists
basecamp_create_todolist - Create a todo list in a todo set, with an optional description
basecamp_update_todolist - Rename a todo list or a group (section), or edit its description
basecamp_move_todolist - Move a todo list to the top, the bottom, or before/after another list
basecamp_create_todolist_group - Create a group (section) in a todo list, at the bottom or at a given place
basecamp_move_todolist_group - Move a group (section) among the groups of its todo list
basecamp_list_todos - List todos in a list, with their groups, and status filtering (active/archived)
basecamp_create_todo - Create new todo with optional description
basecamp_update_todo - Update a todo's title, description, due date, or assignees
basecamp_move_todo - Move a todo to the top, the bottom, or before/after another todo, also into a different list or group
basecamp_reorder_todos - Set the order of all the todos in a list or group, for example to sort them by due date
basecamp_complete_todo - Mark todo as complete
basecamp_uncomplete_todo - Mark todo as incomplete
basecamp_list_comments - List comments on any resource (works universally on all recording types)
basecamp_create_comment - Add comment to any resource
basecamp_update_comment - Update comment with advanced content editing (supports full replacement, append, prepend, search/replace)
basecamp_get_me - Get personal information for the authenticated user
basecamp_list_people - List all people with optional filtering by name, email, or title
basecamp_get_person - Get person details
basecamp_list_kanban_columns - List all columns in a kanban board
basecamp_list_kanban_cards - List cards in a column with steps and assignees
basecamp_get_kanban_card - Get complete details of a specific card
basecamp_create_kanban_card - Create new card with title, content, and optional checklist steps
basecamp_update_kanban_card - Update card with advanced content editing (supports full replacement, append, prepend, search/replace, plus title, due date, assignees, notifications, and complete step array management)
basecamp_move_kanban_card - Move a card to a different column and/or position
basecamp_list_recordings - Browse recent activity globally or across specific projects, with filtering by type, date range, person, and text search. All filters support multiple values for OR-matching (e.g., multiple project IDs, person IDs, types, or search terms)
basecamp_list_campfire_messages - Browse chat messages from Campfires with filtering by campfire, person, text content, and date range. All filters support multiple values for OR-matching
basecamp_list_vaults - List sub-vaults (folders) under a parent vault
basecamp_get_vault - Get a vault's details, including document/upload/sub-vault counts
basecamp_create_vault - Create a new vault (folder)
basecamp_update_vault - Rename a vault
basecamp_list_documents - List documents in a vault, with optional title/content filtering
basecamp_get_document - Get a document's full HTML content
basecamp_create_document - Create a new document (active or draft)
basecamp_update_document - Update a document with advanced content editing (supports full replacement, append, prepend, search/replace)
basecamp_list_uploads - List files uploaded to a vault
basecamp_get_upload - Retrieve an uploaded file: images are returned inline, text files as text, other binary formats saved to disk
basecamp_download_blob - Download an inline <bc-attachment> attachment referenced in document/message/comment HTML content
basecamp_create_attachment - Upload a local file or a remote URL and get the attachable_sgid to embed it with <bc-attachment>
basecamp_get_questionnaire - Get a project's check-ins container
basecamp_list_questions - List automatic check-in questions with schedule and answer counts
basecamp_get_question - Get a single check-in question
basecamp_list_answers - List answers to a check-in question
basecamp_get_answer - Get a single check-in answer
basecamp_create_answer - Post a new answer to a check-in question
basecamp_trash - Move any item (todo list, group, todo, message, comment, document, folder, upload, card) to the trash, with the items in it
basecamp_restore - Bring an item back from the trash or the archive
# Install dependencies
npm install
# Run type checking
npx tsc --noEmit
# Build
npm run build
# Run the live test suite (requires a real, authenticated Basecamp account)
npm test
# Clean build artifacts
npm run cleanMIT
Mark a todo as completed.
Inputs: todo_id (required).
Create a new answer for a check-in question. Content must be HTML.
Inputs: content (required), group_on (required), question_id (required).
Upload a file to Basecamp and get the attachable_sgid needed to embed it in rich text. Use this when you want to show an image or a video in a message, comment, document, card or todo, and that file is not in Basecamp yet — for example an image at an external URL, or a screenshot on disk. Give either file_path or url. The returned sgid goes into a tag. An sgid is tied to this account and does not expire, but it is only shown once here, so embed it in the same session.
Inputs: url, name, file_path, content_type.
Add a comment to any Basecamp resource (message, todo, card, etc.).
Inputs: content (required), recording_id (required).
Create a new document in a vault.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: title (required), content (required), vault_id (required), status.
Create a new card in a kanban column with optional checklist steps.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: title (required), column_id (required), steps, due_on, +3 more.
Create a new message in a Basecamp message board.
Inputs: subject (required), message_board_id (required), status, content, +1 more.
Create a new todo item in a todo list.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: title (required), todolist_id (required), due_on, notify, +3 more.
Create a new todo list in a todo set. Get the todoset_id from the project dock (basecamp_get_project).
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: name (required), todoset_id (required), description.
Create a group (section) in a todo list, to split its todos. The new group goes at the bottom of the groups, or at the place that placement tells. Then create todos in the group with basecamp_create_todo (give the group ID as todolist_id), or move todos into it with basecamp_move_todo (destination_id). To rename the group or to give it a description, use basecamp_update_todolist with the group ID.
Inputs: name (required), todolist_id (required), placement, relative_to_id.
Create a new vault (folder) under a parent vault.
Inputs: title (required), parent_vault_id (required).
Download an inline attachment from a tag found in document/message/comment HTML content. Extract the blob_id and filename from the href attribute (format: https://storage.3.basecamp.com/{accountId}/blobs/{blobId}/download/{filename}). For images, returns the image content that the LLM can see directly. For text-based files, returns the file content as text.
Inputs: blob_id (required), filename (required), content_type.
Get a single check-in answer by its ID.
Inputs: answer_id (required).
Retrieve a single document with its full content.
Inputs: document_id (required).
Get all details of a specific kanban card.
Inputs: card_id (required).
Get your full profile for the authenticated user (id, name, email, title, attachable_sgid). To simply check whether you're logged in, use basecamp_whoami.
Inputs: no named parameters.
Retrieve a single message from a Basecamp message board.
Inputs: message_id (required).
Get details about a specific person.
Inputs: person_id (required).
Fetch detailed information about a specific Basecamp project. This tool retrieves complete project details including name, description, dock configuration, and metadata.
Examples:
Use when: "Get details for project 12345"
Use when: Need full project information including dock configuration
Inputs: project_id (required).
Get a single automatic check-in question with its schedule and metadata.
Inputs: question_id (required).
Get the questionnaire (check-ins container) for a project. Returns the number of questions and their URL.
Inputs: questionnaire_id (required).
Get todo set container for a project. Returns todo lists and groups.
Inputs: todoset_id (required).
Get a file uploaded to a vault. For images, returns the image content that the LLM can see directly. For text-based files (plain text, CSV, JSON, XML, etc.), returns the file content as text. For other binary formats, returns metadata only.
Inputs: upload_id (required).
Get details of a vault (folder) including document/upload/sub-vault counts.
Inputs: vault_id (required).
List answers to a specific check-in question. Returns each answer's content, author, and check-in date.
Inputs: question_id (required).
Browse chat messages from Basecamp Campfires. Campfires are real-time chat rooms within projects.
Use this tool to:
See recent chat activity across all campfires or specific ones
Find messages from specific people
Search message content for keywords
Review chat history since a specific date or time period
All filters support multiple values for OR-matching.
Examples:
"What's been discussed in chat today?" → since: "today"
"Show messages from Alice and Bob" → person_ids: [111, 222]
"Find chat messages mentioning deploy or release" → query: ["deploy", "release"]
"Recent messages in campfire 12345" → campfire_ids: [12345]
Inputs: limit, query, since, person_ids, +1 more.
List comments on any Basecamp resource (message, todo, card, etc.). Works universally on all recording types.
Inputs: recording_id (required).
List documents in a vault.
Inputs: vault_id (required), filter.
List cards in a kanban column.
Inputs: column_id (required).
List all columns in a kanban board.
Inputs: card_table_id (required).
List messages in a Basecamp message board (a single project). For cross-project or time-based browsing across content types, use basecamp_list_recordings instead.
Inputs: message_board_id (required), filter.
List available message types/categories for a Basecamp project
Inputs: bucket_id (required).
List all people in the Basecamp account.
Inputs: filter.
List all projects visible to the authenticated user in a Basecamp account. This tool returns active projects with their IDs, names, descriptions, and metadata. Use this to discover project/bucket IDs needed for accessing messages, todos, and other resources.
Inputs: filter.
List all automatic check-in questions in a questionnaire. Returns each question's title, schedule, paused status, and answer count.
Inputs: questionnaire_id (required).
Browse recent activity across Basecamp by listing recordings. Recordings represent all content in Basecamp: todos, messages, documents, comments, uploads, and more.
Use this tool to:
See what's been happening across all projects or specific projects
Find recent activity by one or more people
Review changes since a specific date or time period
Filter activity by content type (todos, messages, documents, etc.)
Search activity by title text
When to use this vs. the per-resource list tools: use the per-project list tools (basecamp_list_messages, basecamp_list_todos, basecamp_list_documents, basecamp_list_comments, basecamp_list_kanban_cards) to browse items WITHIN a single project; use basecamp_list_recordings for CROSS-project, time-based, or multi-type activity browsing.
All filters support multiple values for OR-matching.
Examples:
"What happened in the last 24 hours?" → since: "24h"
"Show recent todos in project 12345" → project_ids: [12345], type: ["todo"]
"What did Alice and Bob do this week?" → person_ids: [111, 222], since: "7d"
"Find messages mentioning launch across projects 1 and 2" → project_ids: [1, 2], type: ["message"], query: ["launch"]
"Find items about design or UX" → query: ["design", "UX"]
"List all messages across projects" → type: ["message"]
Inputs: sort, type, limit, query, +5 more.
List todos in a todo list. Filter by status: 'active' or 'archived'. The todos come in display order: first the todos that are not in a group, then the todos of each group. The response also lists the groups (sections) of the todo list, also the empty groups.
Inputs: todolist_id (required), status, completed.
List files uploaded to a vault in the Docs & Files section.
Inputs: vault_id (required), filter.
List sub-vaults (folders) under a parent vault in the Docs & Files section.
Inputs: parent_vault_id (required), filter.
Authenticate with Basecamp via OAuth. Opens a browser window for authorization. If you have multiple Basecamp accounts, call first without account_id to see the list, then call again with the desired account_id.
Inputs: account_id.
Remove stored Basecamp credentials.
Inputs: no named parameters.
Move a kanban card to a different column and/or position within that column.
Inputs: card_id (required), column_id (required), position.
Move a todo to a different place: in its todo list, into a group (section), or into a different todo list.
Examples:
Put a todo at the top of its list: placement "top".
Put a todo after another todo: placement "after" and relative_to_id. If the other todo is in a different list or group, the todo moves there.
Move a todo to the bottom of another list or group: placement "bottom" and destination_id.
Only the todos that are not complete have a position. To set the order of many todos, use basecamp_reorder_todos.
Inputs: todo_id (required), placement (required), destination_id, relative_to_id.
Move a todo list to a different place among the todo lists of its project. Get the todo list IDs from basecamp_get_todoset. To move a group (section), use basecamp_move_todolist_group.
Inputs: placement (required), todolist_id (required), relative_to_id.
Move a group (section) to a different place among the groups of its todo list. The groups always show after the todos that are not in a group. Get the group IDs from basecamp_list_todos. You cannot move a group to a different todo list.
Inputs: group_id (required), placement (required), relative_to_id.
Set the order of all the todos in a todo list or in a group (section). Give the IDs of all the todos that are not complete, in the new order. Use this to sort todos, for example by due date. To move one todo, use basecamp_move_todo. The todos in a group are not part of the todo list: to reorder them, give the group ID.
Inputs: todo_ids (required), parent_id (required).
Bring an item back from the trash or from the archive. The items in it come back too, for example the todos of a todo list. To find the IDs of trashed items, use basecamp_list_recordings with status "trashed".
Inputs: recording_id (required).
Move an item to the trash: a todo list, a group (section), a todo, a message, a comment, a document, a folder, an upload or a kanban card. The items in it go too, for example the todos of a todo list. Basecamp deletes the items in the trash permanently after some time. Until then, basecamp_restore brings the item back with the same ID.
Inputs: recording_id (required).
Mark a todo as incomplete (undo completion).
Inputs: todo_id (required).
Update a comment. Use partial content operations when possible to save on token usage.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: comment_id (required), content, content_append, search_replace, +1 more.
Update a document. Use partial content operations when possible to save on token usage.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: document_id (required), title, content, content_append, +2 more.
Update a kanban card including its steps. At least one field (title, content, partial content operations, or steps) must be provided. Use partial content operations when possible to save on token usage.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: card_id (required), steps, title, due_on, +5 more.
Update a message. Use partial content operations when possible to save on token usage.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: message_id (required), content, subject, content_append, +3 more.
Update a todo item. Use partial content operations when possible to save on token usage.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: todo_id (required), title, due_on, notify, +6 more.
Update the name or the description of a todo list, or of a group (section) in a todo list. Use partial content operations on the description when possible to save on token usage. A todo list is complete when all its todos are complete, so this tool cannot complete it.
HTML rules for content:
Allowed tags: p, span, h2, h3, h4, br, strong, em, strike, code, a (with href attribute), pre, ol, ul, li, blockquote, mark, figure, figcaption, table, tbody, tr, th, td, div, bc-attachment.
Wrap every paragraph in ..., and put between one paragraph and the next. Both parts are necessary: newlines and blank lines in your HTML are discarded, and Basecamp shows no space around a on its own, so two paragraphs with nothing between them are shown as one block of text. Do not rely on line breaks in your HTML to separate paragraphs.
Headings: use , , as appropriate.
Inline code: text. Preformatted blocks: text.
Ordered lists: .... Unordered: ....
Tables: Heading...Cell...
To mention a person: . Put the tag where the name must show in the text, for example "Thanks for the fix". The person gets a notification. If the ID is not a person, the tool returns an error and writes nothing. Get person IDs from basecamp_list_people, or from the people in other responses (assignees, creators).
Single image:
Image gallery: wrap multiple in a .
A bc-attachment needs an attachable_sgid. For a file that is already in Basecamp, take the sgid from the HTML content of the message, comment or card that shows it, or use basecamp_list_recordings. For a file that is not in Basecamp yet, such as an image at an external URL or a screenshot on disk, upload it with basecamp_create_attachment and use the sgid that it returns. There is no tag, so an external image URL can never be embedded directly: upload it, or accept that it stays a plain link.
Basecamp auto-enriches bc-attachment tags after saving (adds url, href, filename, content-type, etc.) — you never need to write those.
When you see an existing, already-enriched tag (e.g. from a previous list/get call), leave its inner HTML alone. Before any content_append/content_prepend/search_replace runs, it is automatically collapsed back to its minimal form (sgid, presentation, caption, and content-type for mentions) — you don't need to strip it yourself, and doing so manually is unnecessary and risks mismatched find strings.
Background highlights: ...
Text color highlights: ...
For both, N is 1 (yellow), 2 (amber), 3 (red), 4 (pink), 5 (purple), 6 (blue), 7 (teal), 8 (near-white), or 9 (light gray).
Inputs: todolist_id (required), name, content, content_append, +2 more.
Update the title of a vault (folder).
Inputs: title (required), vault_id (required).
Check login state: show whether you're authenticated and, if so, the basic Basecamp user + account id. For your full profile (id, title, attachable_sgid) use basecamp_get_me.
Inputs: no named parameters.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 57 tools
Each tool targets a distinct resource+action, and the descriptions explicitly differentiate overlapping cases (e.g. list_recordings vs per-project list tools, move_todo vs reorder_todos, complete_todo vs uncomplete_todo). A few conceptual overlaps remain (get_upload vs download_blob, list_documents vs list_recordings), but they are clearly scoped.
Every tool follows the same basecamp_verb_noun convention (basecamp_create_message, basecamp_update_todo, basecamp_list_projects, basecamp_move_kanban_card, basecamp_trash, basecamp_restore). Verbs and resource nouns are used predictably throughout, with no mixing of conventions.
57 tools is well beyond the typical 3-15 sweet spot and is heavy even for a broad domain like Basecamp. To their credit the tools are not redundant and each covers a real resource/operation, but the surface is large enough to strain discovery.
Coverage spans the full Basecamp domain: projects, people, messages, todos/todolists/groups, vaults, documents, uploads, kanban, check-ins (questionnaires/questions/answers), campfire, recordings, comments, attachments, plus auth. CRUD is complete for each area, with trash/restore handling deletion and dedicated tools for ordering and moving items.
Basecamp's official server: projects, to-dos, cards, messages, docs, schedules, and Campfire.
Manage Redbooth tasks, task lists, comments, tags and files from Claude and other AI assistants.
Remote MCP for Kanban AI boards—manage projects, tasks, and comments from AI tools.
Manage projects, tasks, time tracking, and team collaboration through natural language.
