List Gmail threads.

Read one attachment, by the attachmentId messages_get or threads_get lists. A text file comes back as text; a binary one is reported, not decoded.

Create a file's metadata: a folder (mimeType application/vnd.google-apps.folder), a Google Doc / Sheet / Slide, or an empty blob. Media upload is not expressed here. parents takes at most one folder ID.

Appends values to a spreadsheet.

Returns a range of values from a spreadsheet.

List events matching a given search filter.

Create a calendar event; returns details of the event.

Send an email via the Gmail API.

Read one meeting note: its AI summary, the people who attended, the calendar event it was taken against, and the folders it belongs to. Pass include='transcript' to get the transcript inline as well. A transcript too large to inline answers 413 TRANSCRIPT_TOO_LARGE. Read it with notes_transcript_get instead.

Update a page's property values, icon, cover, or trash state. Properties not named are left alone, and a value set to null is cleared. This cannot move a page — use pages_move.

Send a message to a Slack channel, private group, or DM. Provide text for a plain message; set thread_ts to reply inside an existing thread.

Search repositories with qualifiers, e.g. 'topic:cli language:go stars:>500'. Rate limited to 30 requests per minute.

List a repository's languages with bytes of code each — the quickest way to find out what a repository is written in before reading any of it. Answers with an object keyed by language, so there is nothing to page.

Search commits by message, author or date across GitHub — 'repo:owner/name fix flaky test'. The way to find where a change was introduced. Rate limited to 30 requests per minute.

Add people to a channel. With force set, Slack adds everyone it can and reports the rest alongside a success, so read the result.

Add a member to a team.

Create a permission for a file or shared drive. type is user, group, domain or anyone; role is owner, organizer, fileOrganizer, writer, commenter or reader. For user or group, send emailAddress. Concurrent permission writes on the same file are not supported.

Create a page — as a subpage of another page, or as a row of a database by giving its data_source_id as the parent. Content comes as a markdown string Notion parses into blocks, or from a template: one or the other, never both.

Create a new form from a title. Only the title and the document title are honoured: the form is created with no description, no items and default settings. To add questions, call this and then forms_batch_update with the returned formId. Pass unpublished=true to create a form that does not yet accept responses.

List a form's submitted responses, newest page first, up to 5000 per page. The only supported filter is on submission time: pass filter='timestamp >= 2026-01-01T00:00:00Z' to read what has arrived since a point in time. Answers come back keyed by questionId.

Create a blank document with a title. Only the title is honoured — the document is created empty. To add content, call this and then documents_batch_update with the returned documentId.

Get a page's properties. This does not return the page's content — the blocks are read with blocks_children_list, or all at once as Markdown with pages_retrieve_markdown. A property holding more than 25 entries comes back truncated; read it in full with pages_retrieve_property_item.

Create a copy of a file and apply any requested updates with patch semantics. Typically send a new name and optionally parents.

Apply a list of edits to a document. Each request sets exactly one kind of edit — insert_text, replace_all_text, update_text_style, insert_table and so on. Edits apply in order against a document that shifts as they do: when inserting at several indices, order the requests back-to-front so earlier insertions do not move the positions later ones refer to.

Save an email draft to Gmail.

Execute a real-time web search optimized for AI agents. Use when sources are unknown or current web context is needed. Prefer search_depth advanced with chunks_per_source 3 for stronger evidence per source.

Scrape a single URL and optionally extract information. Use when the user wants to read or summarize a specific webpage. Supports markdown, HTML, screenshots, and structured JSON extraction.

Connect one app this server uses. For an app that issues keys, says where to get one and the terminal command that stores it. For Google, once the user's own OAuth client is set, starts the browser sign-in and returns at once: the user approves in the browser and the next call works. To see which apps are connected, call connection_status. Never ask the user for a key in the chat.

See which apps this server is connected to, and how to connect each one that is not. Changes nothing.