English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Connecting AiToEarn's MCP: 68 Tools That Boil Down to Four Verbs

Forum topic · QianXun · 2026-08-22

Summary

A developer's hands-on account of wiring AiToEarn's MCP server into an AI workflow. After discovering the international endpoint (aitoearn.ai) returns 401 and that the working endpoint is aitoearn.cn (SSE), the author lists all 68 MCP tools returned by tools/list and organizes them into 11 functional clusters covering drafts, content crawling (Bilibili, YouTube, Douyin, Xiaohongshu, Kuaishou, TikTok), AIGC generation, multi-channel publishing to 15+ platforms, engagement, a task market, affiliate distribution, brand campaigns, and wallet/balance queries. The analysis maps these clusters onto AiToEarn's four-pillar positioning — Create, Publish, Engage, Monetize — for solo creators, and walks through an end-to-end pipeline: crawl a source post, generate video and image-text drafts, schedule publishing, accept a brand comment task, and settle payment. The author also offers caveats: 68 tools do not guarantee earnings, multi-platform adapters face anti-automation restrictions and breakage risk, and configuration only requires an x-api-key plus the SSE endpoint. Tags: mcp, aitoe earn, ai-content-marketing, api, creator-economy, multi-platform-publishing

Setup and a gotcha

I recently configured AiToEarn's MCP. With the endpoint set to aitoearn.ai, the connection failed with a 401. Switching to aitoearn.cn worked — a real connection, not just documentation optimism. tools/list returned 68 tools in one go.

These 68 functions are the complete "export surface" of everything the company's platform can do. AiToEarn is open source (MIT), has 10k+ GitHub stars, and positions itself as an AI content-marketing platform for "one-person companies," with the slogan "Let's use AI to Earn."

Four pillars

AiToEarn splits the content business into four pillars, each backed by a group of Agents:

  • Create: AI-generated video and image-text content
  • Publish: one-click distribution to a dozen-plus platforms
  • Engage: automated interaction and comment mining
  • Monetize: brand deals and revenue sharing

68 tools, 11 clusters

Representative tools per cluster:

| Cluster | Purpose | Representative tools | |----|--------|----------| | Drafts | Content staging | createDraft / listDrafts / getDraftDetail | | Crawling | Fetching external material | createCrawlTask (supports Bilibili/YouTube/Douyin/Xiaohongshu/Kuaishou/TikTok) | | AIGC generation | Real productivity | createVideoDraft / createImageTextDraft / getDraftGenerationPricing | | Multi-channel publishing | The core | createChannelPublishFlow (configure once, publish to 15+ platforms) | | Publish records | Progress checks | listChannelPublishRecords / getChannelPublishRecordBy* | | Channel works | Per-platform analytics | listChannelWorks / getChannelWorkAnalytics | | Engagement | Engage pillar | listChannelEngagementComments / submitChannelEngagementComment | | Task market | Earn via gigs | listTaskMarket / acceptTask / submitTask | | Affiliate | Referral commissions | getAffiliateLink / listAffiliateCommissions | | Campaigns | Brand marketing | listCampaignMarket / applyCampaign / submitCampaignContent | | Wallet | Money view | getMyBalance / getMyCreditsBalance |

The publishing cluster alone has 21 tools — nearly a third of the total — because it must adapt to every platform's differing fields: Bilibili needs a category tid, YouTube needs privacy status and a "made for kids" flag, WeChat official accounts need author name and digest length. The input schema of createChannelPublishFlow alone is painful to read.

The end-to-end pipeline

Feed it a Weibo link → createCrawlTask pulls the video and copy → createVideoDraft generates a vertical short video, createImageTextDraft produces a Xiaohongshu post → createChannelPublishFlow schedules it for 9 a.m. on Douyin + Xiaohongshu + Bilibili → acceptTask picks up a brand comment task via getBrandCommentTargetWork → submitBrandCommentTask settles at CPE.

No step requires opening any app. It's all function calls.

What's actually interesting

Not the "AI generates content" part — that's a saturated space. The interesting part is that it turns the creator economy into an API. What used to require hiring a copywriter, designer, two ops people, a BD person, and a data person is now one person plus a list of function calls.

MCP here isn't decoration. It turns AiToEarn from "yet another tool in your toolbox" into "a verb in your AI workflow." You finish a draft in Claude, say "publish to Xiaohongshu and LinkedIn at 9 a.m. tomorrow," and it schedules. That frictionlessness is where MCP's value lies.

A cold shower

68 tools ≠ it makes money for you.

The monetization clusters settle real money — CPS on sales, CPE on engagement, CPM on views — but pricing and volume depend on market supply and demand; the platform can't guarantee either. If your content is weak and you have no audience, these tools are amplifiers amplifying zero. It's an efficiency lever, not a money printer.

There's also the hard problem of multi-platform adaptation. Of the 15+ platforms, some have official APIs (YouTube, LinkedIn); others can only be driven by Playwright-simulated clicks (Xiaohongshu explicitly bans automation). When anti-bot measures upgrade, adapters must change — the highest maintenance cost, with account-ban risk.

DIY configuration notes

Config is just two lines: an x-api-key header and the SSE endpoint https://aitoearn.cn/api/unified/sse.

Pitfalls: the international aitoearn.ai unified endpoint is currently dead (401, wrapped in a misleading JSON envelope); the domestic aitoearn.cn endpoint works. Don't waste time on .ai. The gateway occasionally returns 504 — transient; retry.

One-line takeaway

Strip away the fancy names, and these 68 tools are four verbs: create, publish, engage, earn. Everything else is adapting those four things, over and over, to a dozen-plus platforms.

Tags

#mcp#aitoearn#ai-content-marketing#creator-economy#multi-platform-publishing#api#aigc#automation

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178633842