> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sydium.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Drive Sydium from your AI agent over MCP

Connect an AI agent to Sydium and run your accounts by talking to it - draft in your brand voice, publish, schedule, cancel, and pull analytics.

The connect is one browser click: no API keys to paste, no developer app to register. Sydium's MCP server is a real OAuth 2.1 server with dynamic client registration, so the agent registers itself and you just approve the scopes. Works with Claude, ChatGPT, Cursor, or any MCP client.

**Endpoint**

```
https://api.sydium.com/v1/mcp
```

## How it works

Your agent gets **7 tools**. Connecting takes three steps - nothing to install:

<Steps>
  <Step title="Authorize once">
    Point your client at the Sydium MCP endpoint and click **Authorize** in the browser. No developer app to register, no API keys to paste - Sydium registers your client automatically, and you approve exactly which scopes it gets.
  </Step>

  <Step title="Your agent discovers the tools">
    On connect, the agent reads the 7 Sydium tools and their schemas - so it knows precisely what it can do and how to call each one.
  </Step>

  <Step title="Command in plain language">
    Ask for what you want - *"publish this everywhere,"* *"schedule the recap for Friday."* The agent calls the right tools and reports back. You keep the final say, always within the scopes you approved.
  </Step>
</Steps>

## Available tools

| Tool                        | Scope               | What it does                                         |
| --------------------------- | ------------------- | ---------------------------------------------------- |
| `list_accounts`             | `accounts:read`     | List your connected social accounts                  |
| `get_platform_capabilities` | `capabilities:read` | Per-platform posting limits, formats, and post types |
| `generate_caption`          | `generate:write`    | Draft a caption in your brand voice                  |
| `publish_post`              | `posts:write`       | Publish now or schedule to one or more platforms     |
| `get_post_status`           | `posts:read`        | Per-target status of a publish or schedule operation |
| `cancel_post`               | `posts:write`       | Cancel a still-scheduled operation                   |
| `get_analytics`             | `analytics:read`    | Account analytics time-series                        |

<Card title="Tools reference" icon="wrench" href="/mcp/tools">
  Full input schema and behavior for every tool.
</Card>

## Authentication

Your client connects over OAuth: it sends you to a Sydium consent screen, you pick **Authorize**, and the agent gets a scoped token. The agent never sees a password, and it can never do more than the scopes you grant - or more than your own team role allows. The [Client setup](/mcp/connect) guide has the per-client steps and a walkthrough of the consent screen.

## Connect your client

<Card title="Client setup" icon="plug" href="/mcp/connect">
  Step-by-step for Claude Desktop, ChatGPT, and Cursor.
</Card>

## Quick example

Once connected, just talk to your agent:

```
Publish my launch post to all my accounts in my voice,
then schedule the recap thread for Friday at 9am.
```

The agent calls `list_accounts`, `generate_caption`, and `publish_post` (with `scheduleAt`), then reports back. You keep the final say.

## FAQ

<AccordionGroup>
  <Accordion title="Do I need an API key?">
    No. MCP uses one-click browser OAuth with dynamic client registration. API keys are only for the REST API and CLI.
  </Accordion>

  <Accordion title="Which clients are supported?">
    Any MCP-compatible client - Claude (desktop and web), ChatGPT, Cursor, and others. See [Client setup](/mcp/connect).
  </Accordion>

  <Accordion title="Can the agent post videos and images?">
    The agent works with text and captions out of the box. Attaching media through the agent is limited today - media is uploaded and managed in the Sydium app, since agents cannot stream binary uploads. Broader agent-driven media support is planned.
  </Accordion>

  <Accordion title="What can it access?">
    Only the scopes you approve on the consent screen, and never more than your own team role allows.
  </Accordion>
</AccordionGroup>