The social media scheduling API that treats developers as first-class users

One API instead of seventeen

Building against social platform APIs directly means 17 OAuth flows, 17 media pipelines, and 17 flavors of rate limiting. The posterly API collapses that into one surface: you connect accounts once in the dashboard, then create a post on any of them with a single JSON request (one request per account, or use the batch endpoint for several at once). posterly handles token refresh, media transcoding, retries, and publishing infrastructure.

Authentication and pricing

Generate an API key from the dashboard under Settings, in the API section, and send it as Authorization: Bearer pst_live_... on every request. API access starts at $3/month on individual plans, with Agency API at $29/month. Key allowances scale by tier, from 1 key on Starter to 4 on Agency.

Your first request

Create a post with POST https://www.poster.ly/api/v1/posts: pass caption, account_id (from GET /api/v1/accounts), optionally media_urls, and optionally scheduled_at as an ISO 8601 timestamp. Without a schedule the post publishes immediately; with one, posterly's queue publishes it at the right moment with automatic retries. The full endpoint catalog, with runnable examples, lives in the interactive API reference at /reference.

Media that just works

Pass image or video URLs in the create-post body and posterly fetches, validates, and transcodes them per platform: aspect ratios for Instagram, duration caps for X, size limits for TikTok. You upload once and stop thinking about per-platform media specs.

Webhooks for the return path

Register endpoints with POST /api/v1/webhooks and subscribe to lifecycle events: post.created, post.publishing, post.published, post.failed, and more. Every delivery is signed with HMAC SHA-256 in the X-posterly-Signature header so you can verify authenticity before acting on it. This is also the trigger mechanism for Zapier, n8n, and Make workflows.

MCP server for AI agents

If your integration is an AI assistant rather than a backend, use the posterly MCP server: a hosted HTTP endpoint plus an npm stdio package, exposing dozens of tools for Claude, ChatGPT, Cursor, and any MCP-capable client. Your agent schedules posts by talking, not by templating HTTP calls. Details and setup live at /mcp.

CLI for scripts and CI

The posterly CLI wraps the same API for shell scripts and pipelines: post from a GitHub Action when you cut a release, schedule a week of content from a cron job, or generate AI captions from the terminal. Install it from npm and authenticate with the same API key.

Limits and good citizenship

Create-post requests are capped at 100 per hour per key, with separate buckets for media uploads and reads. Responses use standard HTTP status codes with structured JSON errors, so backoff logic is straightforward. If you are building something that needs more headroom, get in touch.