# Social Marketing Workflow Automation  with Soika AI Agent

### **1. Use Case Overview**

**Objective:**\
Enable social media, marketing, or support teams to automate and manage multi-platform social media workflows — including content approval, scheduling, analytics extraction, and task delegation — using a Social Workflow Agent powered by Soika. 

**Primary Actors:**

* Social Media Manager / Marketing Team
* Soika Social Workflow Agent
* Content Approvers / Stakeholders
* Tools: Social APIs (via Webhooks or integrations)

**Target Platforms:**

* Facebook / LinkedIn / Twitter (via integrations)

### **2. Business Value Proposition**

The Soika Social Workflow Agent accelerates the content-to-publish pipeline, reduces dependency on manual reviews, and centralizes tracking for social campaigns. From drafting posts to scheduling and getting manager approvals, all steps are handled automatically through a conversational interface or simple triggers.

#### **Benefits:**

* 70% faster content workflow
* Auto-scheduling and publishing readiness
* Zero-code campaign handling for small teams

### **3. Step-by-step Process**

### **A. AI Agent Creation**

1. Launch Soika Mockingjay platform
2. Prompt:

   > “Create a Social Workflow Agent for Twitter that handles tweet creation, approval, scheduling.”
   >
   > <p align="center"></p>

<figure><img src="/files/hbdsQbUctyXy6BY4BZzZ" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/B8xCO5ZTuaZyfUONN0nZ" alt=""><figcaption><p>Social Media Workflow  Agent is created</p></figcaption></figure>

### **B. Create the Workflow with Nodes and Tool Integration (Twitter )**

### **Step 1: Input field to the Start Node**

#### **Click on the Agent -> Click on the Start Node ->Add the Input field as Paragraph Along with the Variable Name.**

<figure><img src="/files/6ohuz5HR48M3izz5XmTe" alt=""><figcaption></figcaption></figure>

### **Step 2: Add LLM node**

#### **Click on the + icon and Add LLM block. Click on the LLM block and Provide the System Instructions.**

**Connect the Start Node with the LLM Node. Choose your own LLM Model (example llama3.3)**

<figure><img src="/files/PUoBDY25TYTXTS2zXq9y" alt=""><figcaption><p>Click on the + icon and add LLM</p></figcaption></figure>

<figure><img src="/files/JZdjR4HHNdMNxj4uOv15" alt=""><figcaption></figcaption></figure>

### **Step 3: Add System Instruction**

**Example :Choose your Model and Add System Instructions**

The Tweet should not be more than 200 characters based on the title. Give me a brief explanation of the the {{#start.Tweet#}} .Type / for the Context

### **Step 4: Add X(formally Twitter) Tool**

#### **Click on the + icon and Add X(formally Twitter)  from the Tool Section.**

<figure><img src="/files/AXBDuTkp9ca9eCsDwf0u" alt=""><figcaption><p>Select Post Tweet Under X(formerly Tweet) in the Tools Section</p></figcaption></figure>

### **Step 5: Add the Input Variable as LLM text**&#x20;

#### **In the right panel. Type " / " and click on LLM text.**

<figure><img src="/files/1giKk3oZksGpGtTqxCd5" alt=""><figcaption></figcaption></figure>

### **Step 6: Authorize with the Tools**

* **Twitter API**:

  * Connect using Developer credentials
  * Click on the Help link (How to Get)  and Create the API Secret Key and the API Access Key.
  * Once both the Keys are added. Click on Save.&#x20;
  * It will ask for OTP. Generate OTP by clicking on Helper Link

<figure><img src="/files/I2De75WrGITb5KSJOBsz" alt=""><figcaption></figcaption></figure>

### **Step 7** : **Connect with the End node**

<figure><img src="/files/kApXzsWH60ZSHciEBf0y" alt=""><figcaption><p>Select the Output variable as Post Tweet</p></figcaption></figure>

### **Step 8 : Schedule the Workflow**

#### **If you want to Schedule it. Click on Schedule. Add the Cron expression**

<figure><img src="/files/TAVy3WTvRNIP6OTzDhLb" alt=""><figcaption></figcaption></figure>

For example: Click on helper link (To generate cron expression) -> If you want to schedule the post in every 1min add \*\*\*\*\*

<figure><img src="/files/32s7yrno1SismQt4HlZH" alt=""><figcaption></figcaption></figure>

### **Step 9: Publish and Run.**

<figure><img src="/files/0Z7RDvf3hGvEs7I6PF32" alt=""><figcaption></figcaption></figure>

**Give the Input and click on Run**

<figure><img src="/files/XAd9Wk46rm8DWT4zhAAK" alt=""><figcaption></figcaption></figure>

### Step 10: Output at your twitter account

<figure><img src="/files/Cs6BCwJYwI48fjFjGPJl" alt=""><figcaption></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://soika-labs.gitbook.io/soika-mockingjay/use-cases/social-marketing-workflow-automation-with-soika-ai-agent.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.