Tumblr MCP server for AI agents
Let Claude, ChatGPT or your own agent read and act on Tumblr: 33 typed functions through one MCP server. No Tumblr API key needed.
- Site
- tumblr.com
- Functions
- 33
- Sign-in
- For account actions
- Works with
- MCP, HTTP, Node, Python
Connect
An MCP server for Tumblr alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.
https://api.bowmark.ai/mcp/site/tumblrOne MCP server for Tumblr and every other site Bowmark covers. Your agent looks up what it needs and calls it.
https://api.bowmark.ai/mcpSet it up in your clientBoth need a Bowmark API key. The quickstart gets you one and a first result in a few minutes. Functions that need a login use the person's own account: how sign-in works.
Functions
Open a function to see its arguments, what it returns and how to call it.
findBlogFind a blog by name.Arguments
Name Type Required args{ name: string }Yes Returns
TumblrBlogSummaryinterface TumblrBlogSummary { /** The blog's own identifier, what every other tumblr.com function takes * as a blog argument. */ name: string; title: string; url: string; totalPosts: number; description: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/findBlogCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.findBlog({ name: "staff" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.findBlog({"name": "staff"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/findBlog \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"name":"staff"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.findBlog({ name: "staff" });findTagFind posts by searching a tag.Arguments
Name Type Required tagstringThe tag, with or without a leading #.Yes limitnumber1-20, default 20.No beforenumber | stringA previous page's nextCursor (a unix timestamp, as a number or a string).No Returns
TumblrSearchPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrSearchPage { posts: TumblrPost[]; /** Pass back as cursor for the next page; null on the last page. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/findTagCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.findTag({ tag: "cats", limit: 3 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.findTag({"tag": "cats", "limit": 3}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/findTag \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"tag":"cats","limit":3}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.findTag({ tag: "cats", limit: 3 });getBlogInfoGet metadata for a blog: name, title, url, post count, description, last-post time, NSFW/adult flags, ask and subscribe settings, and avatar.Arguments
Name Type Required args{ name: string }Yes Returns
TumblrBlogInfointerface TumblrBlogInfo { name: string; title: string; url: string; totalPosts: number; description: string; /** ISO timestamp of the blog's last post, or null when the site did not send one. */ updatedAt: string | null; isNsfw: boolean; isAdult: boolean; asksEnabled: boolean; asksAllowAnon: boolean; canSubscribe: boolean; /** The largest avatar image the site published, or "" when none came back. */ avatarUrl: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getBlogInfoCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getBlogInfo({ name: "staff" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getBlogInfo({"name": "staff"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getBlogInfo \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"name":"staff"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getBlogInfo({ name: "staff" });getBlogPostsList posts from a blog, newest first.Arguments
Name Type Required namestringA blog name or url, as findBlog takes.Yes limitnumber1-20, default 20.No cursorstringA previous page's nextCursor.No Returns
TumblrPostPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrPostPage { posts: TumblrPost[]; totalPosts: number; /** Pass back as cursor for the next page; null on the last page. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getBlogPostsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getBlogPosts({ name: "staff", limit: 2 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getBlogPosts({"name": "staff", "limit": 2}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getBlogPosts \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"name":"staff","limit":2}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getBlogPosts({ name: "staff", limit: 2 });getDashboardThe signed-in caller's dashboard feed.Arguments
Name Type Required limitnumberNo cursorstringNextCursor off the previous page.No Returns
TumblrFeedPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrFeedPage { posts: TumblrPost[]; /** Pass back as cursor for the next page; null on the last. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getDashboardCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getDashboard({ limit: 2 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getDashboard({"limit": 2}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getDashboard \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"limit":2}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getDashboard({ limit: 2 });getPostGet a single post by ID with its content and metadata.Arguments
Name Type Required namestringThe blog name, with id; or omit both and pass url.No idstringNo urlstringA post url, e.g. https://staff.tumblr.com/post/822057428507049984.No Returns
TumblrPostinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getPostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getPost({});import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getPost({}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getPost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getPost({});getPostNotesGet notes (likes and reblogs) on a post with the blog names that interacted with it.Arguments
Name Type Required namestringThe blog name, with id; or omit both and pass url.No idstringNo urlstringA post url, e.g. https://staff.tumblr.com/post/822057428507049984.No cursorstringA previous page's nextCursor.No Returns
TumblrNotesPageinterface TumblrPostNote { type: string; blogName: string; blogUrl: string; /** ISO timestamp of the interaction. */ timestamp: string | null; /** The reply's own text, present only when type is "reply". */ replyText: string | null; /** The blog this reblog is from, present only when type is "reblog". */ reblogParentBlogName: string | null; } interface TumblrNotesPage { notes: TumblrPostNote[]; totalNotes: number; totalLikes: number; totalReblogs: number; totalReplies: number; /** Pass back as cursor for the next page; null on the last page. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getPostNotesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getPostNotes({ name: "staff", id: "822057428507049984" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getPostNotes({"name": "staff", "id": "822057428507049984"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getPostNotes \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"name":"staff","id":"822057428507049984"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getPostNotes({ name: "staff", id: "822057428507049984" });getTagTimelineGet posts tagged with a specific tag.Arguments
Name Type Required tagstringThe tag, with or without a leading #.Yes sort"top" | "recent""top" (default) or "recent".No limitnumber1-20, default 20.No cursorstringA previous page's nextCursor.No Returns
TumblrSearchPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrSearchPage { posts: TumblrPost[]; /** Pass back as cursor for the next page; null on the last page. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getTagTimelineCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getTagTimeline({ tag: "cats", limit: 3 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getTagTimeline({"tag": "cats", "limit": 3}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getTagTimeline \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"tag":"cats","limit":3}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getTagTimeline({ tag: "cats", limit: 3 });getTrendingPostsGet posts that are currently trending on Tumblr.Arguments
Name Type Required limitnumber1-20, default 20.No cursorstringA previous page's nextCursor.No Returns
TumblrSearchPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrSearchPage { posts: TumblrPost[]; /** Pass back as cursor for the next page; null on the last page. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getTrendingPostsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getTrendingPosts({ limit: 3 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getTrendingPosts({"limit": 3}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getTrendingPosts \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"limit":3}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getTrendingPosts({ limit: 3 });getTrendingTagsGet tags that are currently trending on Tumblr.Arguments
None.
Returns
TumblrTrendingTagsPageinterface TumblrTrendingTag { tag: string; followerCount: number; recentPostCount: number; relatedTags: string[]; } interface TumblrTrendingTagsPage { tags: TumblrTrendingTag[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/getTrendingTagsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.getTrendingTags();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.getTrendingTags())curl -X POST https://api.bowmark.ai/v1/sites/tumblr/getTrendingTags \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.getTrendingTags();listAsksAsks and submissions waiting in the signed-in caller's own blog inbox.Arguments
Name Type Required blogstringThe signed-in caller's own blog.Yes limitnumberNo offsetnumberNo Returns
TumblrAsksPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrAsksPage { /** Asks and submissions waiting in the inbox, as post rows. */ asks: TumblrPost[]; /** Pass as offset for the next page, or null on the last. */ nextOffset: number | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listAsksCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listAsks({ blog: "staff" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listAsks({"blog": "staff"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listAsks \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"staff"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listAsks({ blog: "staff" });listBlogFollowersList followers of a blog; caller must own the blog.Arguments
Name Type Required blogstringA blog the signed-in caller owns.Yes limitnumber1-20, default 20.No offsetnumberNo Returns
TumblrFollowersPageinterface TumblrFollower { name: string; url: string; /** Whether the signed-in caller follows this blog back. */ following: boolean; updatedAt: string | null; } interface TumblrFollowersPage { totalFollowers: number; followers: TumblrFollower[]; /** Pass as offset for the next page, or null on the last. */ nextOffset: number | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listBlogFollowersCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listBlogFollowers({ blog: "staff" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listBlogFollowers({"blog": "staff"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listBlogFollowers \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"staff"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listBlogFollowers({ blog: "staff" });listDraftsList the drafts or queued posts on the signed-in caller's own blog.Arguments
Name Type Required blogstringThe signed-in caller's own blog.Yes state"draft" | "queue""draft" (default) or "queue".No limitnumberNo offsetnumberNo Returns
TumblrDraftsPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrDraftsPage { state: "draft" | "queue"; posts: TumblrPost[]; /** Pass as offset for the next page, or null on the last. */ nextOffset: number | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listDraftsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listDrafts({ blog: "staff" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listDrafts({"blog": "staff"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listDrafts \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"staff"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listDrafts({ blog: "staff" });listFollowedTagsList the hashtags the signed-in caller follows.Arguments
None.
Returns
TumblrFollowedTagsPageinterface TumblrFollowedTagsPage { /** Tag names without the "#". */ tags: string[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listFollowedTagsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listFollowedTags();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listFollowedTags())curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listFollowedTags \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listFollowedTags();listFollowingList the blogs the signed-in caller follows.Arguments
Name Type Required limitnumberNo offsetnumberNo Returns
TumblrFollowingPageinterface TumblrFollowedBlog { name: string; title: string; url: string; updatedAt: string | null; } interface TumblrFollowingPage { totalBlogs: number; blogs: TumblrFollowedBlog[]; /** Pass as offset for the next page, or null on the last. */ nextOffset: number | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listFollowingCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listFollowing({ limit: 2 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listFollowing({"limit": 2}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listFollowing \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"limit":2}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listFollowing({ limit: 2 });listLikesList the signed-in caller's own likes.Arguments
Name Type Required limitnumberNo cursorstringNextCursor off the previous page.No Returns
TumblrLikesPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrLikesPage { posts: TumblrPost[]; totalLikes: number; nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listLikesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listLikes({ limit: 2 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listLikes({"limit": 2}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listLikes \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"limit":2}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listLikes({ limit: 2 });listMyBlogsList the blogs the signed-in caller owns.Arguments
None.
Returns
TumblrMyBlogsPageinterface TumblrMyBlog { name: string; title: string; url: string; /** True for the caller's primary blog. */ primary: boolean; } interface TumblrMyBlogsPage { blogs: TumblrMyBlog[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listMyBlogsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listMyBlogs();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listMyBlogs())curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listMyBlogs \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listMyBlogs();listNotificationsActivity on the signed-in caller's own blog.Arguments
Name Type Required blogstringThe signed-in caller's own blog.Yes limitnumberPage size, applied to the first page too.No cursorstringNo Returns
TumblrNotificationsPageinterface TumblrNotification { /** like, reblog, follow, reply, ask, mention, Tumblr's own type string. */ type: string; fromBlog: string; postId: string | null; text: string; at: string | null; } interface TumblrNotificationsPage { notifications: TumblrNotification[]; nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/listNotificationsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.listNotifications({ blog: "staff" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.listNotifications({"blog": "staff"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/listNotifications \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"staff"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.listNotifications({ blog: "staff" });searchBlogPostsSearch posts within a specific blog.Arguments
Name Type Required namestringA blog name or url, as findBlog takes.Yes querystringYes limitnumber1-20, default 20.No cursorstringA previous page's nextCursor.No Returns
TumblrSearchPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrSearchPage { posts: TumblrPost[]; /** Pass back as cursor for the next page; null on the last page. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/searchBlogPostsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.searchBlogPosts({ name: "staff", query: "tumblr", limit: 3 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.searchBlogPosts({"name": "staff", "query": "tumblr", "limit": 3}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/searchBlogPosts \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"name":"staff","query":"tumblr","limit":3}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.searchBlogPosts({ name: "staff", query: "tumblr", limit: 3 });searchPostsSearch all posts across Tumblr by keyword or phrase.Arguments
Name Type Required querystringYes limitnumber1-20, default 20.No cursorstringA previous page's nextCursor.No Returns
TumblrSearchPageinterface TumblrPost { /** The post id as a string, what getPost takes. */ id: string; blogName: string; url: string; shortUrl: string; /** ISO timestamp the post was published. */ publishedAt: string | null; tags: string[]; summary: string; /** Every text block, in order, joined by newlines. */ text: string; /** The largest rendition of every image block, in order. */ images: string[]; /** Urls of link, video and audio blocks, in order. */ links: string[]; noteCount: number; likeCount: number; reblogCount: number; isNsfw: boolean; /** The blog this was reblogged from, or null for an original post. */ rebloggedFrom: string | null; reblogKey: string; } interface TumblrSearchPage { posts: TumblrPost[]; /** Pass back as cursor for the next page; null on the last page. */ nextCursor: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/searchPostsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.searchPosts({ query: "cats", limit: 3 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.searchPosts({"query": "cats", "limit": 3}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/searchPosts \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"query":"cats","limit":3}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.searchPosts({ query: "cats", limit: 3 });
createPostCreate a new post on the signed-in account's blog.Arguments
Name Type Required blogstringYour own blog name, or its tumblr.com url.Yes textstringThe post body; blank lines split it into paragraphs.Yes titlestringOptional heading above the body.No tagsstring[]No state"published" | "draft" | "queue" | "private"Default "published".No Returns
TumblrComposeResultinterface TumblrComposeResult { ok: true; /** The post's id, or null when Tumblr's answer carried none. */ id: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/createPostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.createPost({ blog: "<blog>", text: "<text>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.createPost({"blog": "<blog>", "text": "<text>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/createPost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"<blog>","text":"<text>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.createPost({ blog: "<blog>", text: "<text>" });deletePostDelete a post from the signed-in account's blog.Arguments
Name Type Required blogstringThe signed-in caller's own blog.Yes idstringYes Returns
{ ok: true }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/deletePostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.deletePost({ blog: "<blog>", id: "<id>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.deletePost({"blog": "<blog>", "id": "<id>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/deletePost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"<blog>","id":"<id>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.deletePost({ blog: "<blog>", id: "<id>" });editPostEdit an existing post on the signed-in account's blog.Arguments
Name Type Required blogstringYes idstringThe post id on your own blog.Yes textstringReplaces the whole body.Yes titlestringNo tagsstring[]No state"published" | "draft" | "queue" | "private"No Returns
TumblrComposeResultinterface TumblrComposeResult { ok: true; /** The post's id, or null when Tumblr's answer carried none. */ id: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/editPostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.editPost({ blog: "<blog>", id: "<id>", text: "<text>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.editPost({"blog": "<blog>", "id": "<id>", "text": "<text>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/editPost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"<blog>","id":"<id>","text":"<text>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.editPost({ blog: "<blog>", id: "<id>", text: "<text>" });followBlogFollow a blog from the signed-in account.Arguments
Name Type Required namestringA blog name like "staff", or its tumblr.com url.Yes Returns
TumblrEngageResultinterface TumblrEngageResult { ok: true; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/followBlogCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.followBlog({ name: "<name>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.followBlog({"name": "<name>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/followBlog \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"name":"<name>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.followBlog({ name: "<name>" });followTagFollow a hashtag on the signed-in caller's account.Arguments
Name Type Required tagstringThe hashtag, "cats" or "#cats".Yes Returns
TumblrTagResultinterface TumblrTagResult { ok: true; /** The tag as sent, without the "#". */ tag: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/followTagCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.followTag({ tag: "<tag>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.followTag({"tag": "<tag>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/followTag \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"tag":"<tag>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.followTag({ tag: "<tag>" });likePostLike a post from the signed-in account.Arguments
Name Type Required idstringThe post id, off a post row.Yes reblogKeystringThe post row's reblogKey.Yes Returns
TumblrEngageResultinterface TumblrEngageResult { ok: true; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/likePostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.likePost({ id: "<id>", reblogKey: "<reblogKey>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.likePost({"id": "<id>", "reblogKey": "<reblogKey>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/likePost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"id":"<id>","reblogKey":"<reblogKey>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.likePost({ id: "<id>", reblogKey: "<reblogKey>" });reblogPostReblog a post to the signed-in account's blog.Arguments
Name Type Required blogstringThe signed-in caller's own blog to reblog onto.Yes idstringYes reblogKeystringYes commentstringNo Returns
TumblrReblogResultinterface TumblrReblogResult { ok: true; /** The new reblog's own post id, what unreblogPost takes. */ id: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/reblogPostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.reblogPost({ blog: "<blog>", id: "<id>", reblogKey: "<reblogKey>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.reblogPost({"blog": "<blog>", "id": "<id>", "reblogKey": "<reblogKey>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/reblogPost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"<blog>","id":"<id>","reblogKey":"<reblogKey>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.reblogPost({ blog: "<blog>", id: "<id>", reblogKey: "<reblogKey>" });sendMessageSend an ask (public or anonymous) to a blog.Arguments
Name Type Required blogstringYes textstringYes anonymousbooleanAsk without your name attached; the recipient must accept anonymous asks.No Returns
TumblrComposeResultinterface TumblrComposeResult { ok: true; /** The post's id, or null when Tumblr's answer carried none. */ id: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/sendMessageCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.sendMessage({ blog: "<blog>", text: "<text>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.sendMessage({"blog": "<blog>", "text": "<text>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/sendMessage \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"<blog>","text":"<text>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.sendMessage({ blog: "<blog>", text: "<text>" });unfollowBlogUnfollow a blog from the signed-in account.Arguments
Name Type Required namestringA blog name like "staff", or its tumblr.com url.Yes Returns
TumblrEngageResultinterface TumblrEngageResult { ok: true; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/unfollowBlogCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.unfollowBlog({ name: "<name>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.unfollowBlog({"name": "<name>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/unfollowBlog \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"name":"<name>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.unfollowBlog({ name: "<name>" });unfollowTagStop following a hashtag on the signed-in caller's account.Arguments
Name Type Required tagstringThe hashtag, "cats" or "#cats".Yes Returns
TumblrTagResultinterface TumblrTagResult { ok: true; /** The tag as sent, without the "#". */ tag: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/unfollowTagCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.unfollowTag({ tag: "<tag>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.unfollowTag({"tag": "<tag>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/unfollowTag \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"tag":"<tag>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.unfollowTag({ tag: "<tag>" });unlikePostRemove a like from a post on the signed-in account.Arguments
Name Type Required idstringThe post id, off a post row.Yes reblogKeystringThe post row's reblogKey.Yes Returns
TumblrEngageResultinterface TumblrEngageResult { ok: true; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/unlikePostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.unlikePost({ id: "<id>", reblogKey: "<reblogKey>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.unlikePost({"id": "<id>", "reblogKey": "<reblogKey>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/unlikePost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"id":"<id>","reblogKey":"<reblogKey>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.unlikePost({ id: "<id>", reblogKey: "<reblogKey>" });unreblogPostRemove a reblog from the signed-in account's blog.Arguments
Name Type Required blogstringThe signed-in caller's own blog.Yes idstringYes Returns
{ ok: true }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/unreblogPostCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.unreblogPost({ blog: "<blog>", id: "<id>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.unreblogPost({"blog": "<blog>", "id": "<id>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/unreblogPost \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"<blog>","id":"<id>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.unreblogPost({ blog: "<blog>", id: "<id>" });updateBlogInfoChange the title and description of one of the signed-in caller's own blogs.Arguments
Name Type Required blogstringThe signed-in caller's own blog.Yes titlestringNo descriptionstringNo Returns
TumblrUpdateBlogResultinterface TumblrUpdateBlogResult { ok: true; blog: string; title?: string; description?: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/tumblr/updateBlogInfoCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.tumblr.updateBlogInfo({ blog: "<blog>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.tumblr.updateBlogInfo({"blog": "<blog>"}))curl -X POST https://api.bowmark.ai/v1/sites/tumblr/updateBlogInfo \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"blog":"<blog>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.tumblr.updateBlogInfo({ blog: "<blog>" });
Use it from code
- Node
npm i @bowmark/web - Python
pip install bowmark-web - HTTP
POST /v1/sites/tumblr/<function> - OpenAPI
/v1/sites/tumblr/openapi.json