About

One API for social media. Built by a developer.

SocialAPI.ai is an indie project, bootstrapped in France. It started from a simple frustration: building six different integrations just to manage one social media presence is way too much work.

The problem

Every social platform has its own API, its own OAuth flow, its own webhook format, its own rate limits, and its own error codes. If you want to reply to a comment on Instagram, respond to a Google review, and send a DM on Facebook, you need three completely separate integrations with different token refresh logic, different pagination, and different retry strategies.

Most existing tools focus on publishing and scheduling. They push content out. But the real operational work for businesses and AI agents is inbound: reading what people say, triaging it, and responding. No unified API existed for that workflow across social and review platforms.

So I built one.

How it works

One schema, every platform

Instagram calls it a "comment," Google calls it a "review," Facebook calls it a "message." SocialAPI.ai normalizes everything into consistent types with uniform fields and consistent error codes. One integration handles every platform.

42 MCP tools for AI agents

The MCP server was built alongside the REST API because AI agents are a primary use case. Every REST endpoint has a matching MCP tool. Claude, ChatGPT, or any MCP-compatible runtime can connect without custom code.

Publish and engage

Create, schedule, and cross-post content across platforms. Then manage the conversations that follow: reply to comments, answer DMs, respond to reviews, all from the same API.

Webhooks for real-time events

Get notified instantly when someone comments, sends a DM, or leaves a review. Webhooks are delivered over HTTPS with automatic retry and exponential backoff.

Your data is safe

SocialAPI.ai does store data to provide a fast, reliable service. But security is built into every layer, and we only keep what we need.

Encrypted at rest

All OAuth tokens are encrypted with AES-256-GCM. They are decrypted per-request and never cached in plaintext.

Minimal storage footprint

We store connected account metadata, posts, usage logs, and webhook state. Social media content like comments and DMs is fetched in real time from platform APIs and not persisted.

Hashed API keys

Keys are hashed with SHA-256 for lookup and verified with bcrypt. The raw key is shown once at creation and never stored.

HTTPS everywhere

All API traffic is encrypted in transit. Webhook deliveries use HTTPS with automatic retry.

GDPR-aligned

Built and hosted in accordance with European data protection standards. Deleting your account removes all associated data.

Supported platforms

Six platforms are live today. Two more are in active development.

PlatformCapabilitiesStatus
InstagramComments, DMs, mentions, publishingLive
FacebookComments, DMs, mentions, publishingLive
Google Business ProfileReviews, replies, local postsLive
LinkedInComments, publishingLive
ThreadsComments, publishingLive
TikTokPublishingLive
YouTubeCommentsSoon
X / TwitterMentions, DMsSoon

SDKs and community

We provide official SDKs and the MCP server on GitHub. The MCP server is also listed on agent marketplaces including Smithery, Composio, MCP.so, PulseMCP, and mcpmarket.

Bug reports, feature requests, and pull requests on the SDKs are welcome. If you are building something cool with the API, I would love to hear about it.

Say hello

This is an indie project and every email gets a personal reply.