One API for every network: code, no-code, or AI agent
Schedule posts, generate captions, read analytics, track listening, and manage DM automation across 11 networks, from Zapier or Make, your own code, or an AI agent like Claude over MCP. Everything ships through each network's official API.
Open Settings → API keys, name one ("Zapier", "Claude", "Our CMS"), and copy it once. It's shown a single time and stored hashed. Revoke it any moment and it stops working instantly, no effect on your account.
1
API keys
kdz_live_…a1b2 · Zapier
kdz_live_…7f9c · Claude (MCP)
Revoke any time
02
Step 2: Connect three ways, same key
Pick whatever fits. No-code with Zapier, Make or Pipedream. The REST API or the zero-dependency kadenzo CLI from your own systems. Or the MCP server, so an AI assistant can use Kadenzo directly. One key, one account, your choice.
2
Connect through
→No-code: Zapier · Make · Pipedream
→Developers: REST API · CLI
✓AI agents: MCP (Claude · Cursor · ChatGPT)
03
Step 3: Do far more than schedule
It isn't a bare post endpoint. Generate a caption from a topic, upload media, list and edit your queue, read engagement analytics, or ask for the best time to post, then schedule it. The same key also runs your listening (keywords, mention analytics, Share of Voice) and your DM automation rules, funnel stats included.
3
generate → best time → schedule
POST /generate
caption drafted
GET /best-times
Tue · 9:00 AM
POST /posts
queued · X · LinkedIn
04
Step 4: It publishes itself, safely
Runs on its own
At the scheduled minute the post goes live through each network's official API, the same pipeline as a post you scheduled by hand. No password sharing, no scraping, and a key can only ever touch accounts you've already connected.
Published
X · live 9:00 AM
LinkedIn · live 9:00 AM
Official APIs only
What you get
No-code, with Zapier & Make
Connect Kadenzo to thousands of apps through Zapier, Make, and Pipedream. Wire the trigger once and posts schedule themselves. No developer, no maintenance.
Works with your AI agent (MCP)
Drop the Kadenzo MCP server into Claude Desktop, Cursor, or ChatGPT and your agent gets 23 real tools: scheduling, generation, analytics, listening, DM automation. "Write a launch thread, find the best time, and schedule it" just works, no glue code.
Generate, then ship
Ask the API (or your agent) for a caption from a topic, pull the best time to post from your own data, and queue it. The write-it-and-time-it loop, not just a place to drop finished posts.
Analytics you can query
Pull engagement metrics per account or per post, mention analytics for your listening keywords, even a Share of Voice report against competitors, so your own dashboards and agents can read your numbers, not just push posts in.
REST API and a CLI
One documented set of endpoints, key auth, JSON in and out, plus a zero-dependency kadenzo command-line tool for scripts and the terminal. The same validation the in-app composer runs.
Keys you control
Generate as many as you need, label them per tool or agent, see when each was last used, and revoke any one instantly. Keys are stored hashed and shown in full only once.
Kadenzo vs A basic scheduler
A basic scheduler
Schedule from another tool automatically
Sometimes
No-code setup (Zapier / Make / Pipedream)
Sometimes
REST API + command-line tool
Works from an AI agent (MCP)
Generate captions via API or agent
Read analytics via the API
Best time to post from your own data
Manage listening keywords & Share of Voice via API
DM automation rules with funnel stats via API
Publishes through official platform APIs
Varies
Kadenzo
The thinking
Most of the work happens before the post
Scheduling a post is the visible part. The invisible part is everything that feeds it: the spreadsheet your team drops ideas into, the CMS that publishes the blog you want to announce, the approval that finally lands, the conversation with an AI assistant where the caption gets written.
Stop being the wire between your other tools and your posting calendar.
That's the manual work Kadenzo's API removes. Wire it to Zapier, Make, or Pipedream; call the REST API or CLI from your own systems; or let an AI agent drive it over MCP. An approved draft becomes a scheduled post, with no one tabbing between tools to make it happen.
It's the same promise as the rest of Kadenzo, pushed one step earlier: you stop being the wire between your systems and your posting calendar.
When this happensTrigger
“Launch teaser 🚀”
New row added · publish Tue 9:00 AM
from any of your tools
Zapier · Make · API · MCP
Scheduled in Kadenzo
Tuesday · 9:00 AM
Queued
Launch teaser 🚀 Here’s what’s new this week…
Goes live on
Runs automatically
Who it's for
AI builders
Your agent runs your social
Connect the MCP server and ask Claude or Cursor to draft, time, and schedule posts. It generates the copy, picks the best slot from your own data, queues it, and can read back how earlier posts performed, all in one conversation.
Agencies
A client's calendar from your own pipeline
Generate a month of a client's posts in your internal tool and push them all into Kadenzo by API or CLI, then manage, edit, and adjust them on the calendar like any other post.
In-house teams
Announce whatever you ship
Wire your CMS or release tool to Kadenzo so every new article or launch fans out into a scheduled cross-network post the moment it goes live. No code beyond the trigger.
Frequently asked
Do I need to be a developer to use this?
No. The simplest path is no-code: connect Kadenzo to Zapier, Make, or Pipedream, pick a trigger, and posts schedule themselves. If you work through an AI assistant, the MCP server lets Claude, Cursor, or ChatGPT run Kadenzo for you. The REST API and CLI are there if you'd rather write the integration yourself, all optional.
What can the API actually do?
Well beyond scheduling. You can create, list, edit, and cancel posts; upload image and video files; generate a caption from a topic; read engagement analytics per account or per post; ask for your personalized best time to post; and post comments or replies on Instagram, Facebook, and LinkedIn. The listening endpoints read your mentions, manage your tracked keywords, and return mention analytics and a Share of Voice report. The DM endpoints create, edit, and toggle your automation rules, with each rule's funnel (reached, opted in, clicked) and your monthly contact usage in the response. Every action still runs through each network's official API.
Can I read analytics or pull data out by API?
Yes. There are analytics endpoints for a whole account's recent posts and for an individual post, so your own dashboards or an AI agent can read your engagement numbers, not just push posts in. There's also a personalized best-time-to-post endpoint computed from your own history.
Can I use it from an AI agent like Claude?
Yes, that's what the MCP server is for. Add kadenzo-mcp to Claude Desktop, Cursor, ChatGPT, or any MCP client and the agent gets 23 tools: scheduling, caption generation, analytics, listening keywords and Share of Voice, and DM automation rules, using an API key you generate in your settings. There's also a kadenzo CLI for the terminal and scripts.
How does authentication work?
Generate an API key in Settings → API keys. It's shown once and stored hashed, so copy it when you create it, then send it as a Bearer token on each request. Name keys per tool or agent, see when each was last used, and revoke any one instantly without affecting your account.
Is it safe to let another tool or agent post for me?
Yes, within tight limits. A key can only act on accounts you've already connected and only within your plan's limits. It can't reach a channel you don't own, and everything publishes through official platform APIs, never scraping or password sharing. Heavier actions like AI generation and replies have daily caps, and revoking the key stops it working immediately. And there's deliberately no send-a-DM endpoint: the API manages your automation rules, but messages only ever go out through the automation engine and its guardrails.
Which plans include API access?
API access is included on Professional and Enterprise. Endpoints follow your plan from there: personalized best-times is Professional and up, listening endpoints follow your plan's keyword limits, and the Share of Voice report is Enterprise. You generate keys yourself in settings; there's no sales call to unlock it.
