Skip to content

SkillUp Operations Support Bot

An AI-powered support assistant that helps Program Officers, Trainers, and Site Assistants execute the SkillUp Bootcamp with excellence. The bot provides instant access to operational guidance, safeguarding protocols, and retention strategies through Telegram and WhatsApp, backed by a knowledge base of bootcamp documentation.

Purpose

No business context provided yet — add a context.md to enrich this documentation.

Based on the workflow configuration, this bot appears to serve the SkillUp Bootcamp program, supporting Out-of-School Youth (OoSY) in Kenya through a 4-week in-person + 2-week virtual entrepreneurial skills program. The AI agent helps bootcamp staff navigate daily operations, apply safeguarding protocols, deliver quality training experiences, and maintain Educate!'s cultural standards while protecting participant well-being.

How It Works

  1. Document Upload: Staff upload bootcamp handbooks and operational documents via a web form (PDF/CSV files)
  2. Knowledge Processing: Documents are processed and stored in an in-memory vector database with OpenAI embeddings
  3. Message Reception: The bot receives questions via Telegram or WhatsApp (Twilio)
  4. AI Processing: An OpenAI GPT-4o-mini agent processes the query, consulting the knowledge base when needed
  5. Response Delivery: The bot sends contextual, actionable guidance back through the same channel
  6. Memory Management: Conversation history is maintained per chat session for context continuity

Workflow Diagram

graph TD
    A[Upload your file here] --> B[Insert Data to Store]
    B --> C[Default Data Loader]
    C --> D[Embeddings OpenAI]
    D --> E[Query Data Tool]

    F[Telegram Trigger] --> G[AI Agent]
    H[Twilio Trigger] --> G

    E --> G
    I[OpenAI Chat Model] --> G
    J[Simple Memory] --> G

    G --> K[Send a text message]
    G --> L[Send an SMS/MMS/WhatsApp message]

    D --> B
    D --> E

Triggers

  • Form Trigger: Web form for uploading bootcamp documentation (PDF/CSV files)
  • Telegram Trigger: Receives messages from Telegram users
  • Twilio Trigger: Receives WhatsApp/SMS messages via Twilio webhook

Nodes Used

Node Type Purpose
Form Trigger Provides web interface for document uploads
Default Data Loader Processes binary files (PDF/CSV) into text documents
Embeddings OpenAI Creates vector embeddings for document chunks
Vector Store In Memory Stores and retrieves document embeddings (insert/query modes)
AI Agent Main conversational AI with SkillUp-specific system prompt
OpenAI Chat Model GPT-4o-mini language model for generating responses
Buffer Window Memory Maintains conversation context (20 message window)
Telegram Sends responses back to Telegram users
Twilio Sends responses back via WhatsApp/SMS

External Services & Credentials Required

  • OpenAI API: For embeddings and chat completions
    • Credential: "OpenAi account"
    • Used for: Document embeddings and GPT-4o-mini responses
  • Telegram Bot API: For Telegram messaging
    • Credential: "Waringa"
    • Used for: Receiving and sending Telegram messages
  • Twilio API: For WhatsApp/SMS messaging
    • Credential: "Twilio samy a"
    • Used for: Receiving and sending WhatsApp/SMS messages

Environment Variables

No specific environment variables are configured in this workflow. All external service connections use n8n credential management.

Data Flow

Input: - Document files (PDF/CSV) via web form - Text messages from Telegram or WhatsApp/SMS - User queries about bootcamp operations, safeguarding, retention strategies

Processing: - Documents converted to embeddings and stored in vector database - User messages processed by AI agent with access to knowledge base - Conversation history maintained per chat session

Output: - Actionable guidance and responses sent back through original messaging channel - Responses include specific procedures, escalation steps, and cultural considerations - Maintains Educate!'s values: Youth First, Exceed Expectations, Always Learning, One Team Many Views, Startup Mindset

Error Handling

No explicit error handling nodes are present in this workflow. Error handling relies on n8n's default mechanisms and individual node error settings.

Known Limitations

  • Workflow is currently archived and inactive
  • Uses in-memory vector storage (data lost on workflow restart)
  • No persistent conversation history beyond current session
  • Limited to PDF and CSV file formats for document upload
  • No authentication or access control on the web form

No related workflows identified in the current configuration.

Setup Instructions

  1. Import Workflow: Import the JSON configuration into your n8n instance

  2. Configure Credentials:

    • Set up OpenAI API credential with valid API key
    • Configure Telegram Bot credential with bot token
    • Set up Twilio credential with account SID and auth token

  3. Activate Triggers:

    • Enable the Form Trigger and note the webhook URL for document uploads
    • Configure Telegram webhook URL in your bot settings
    • Set up Twilio webhook URL for WhatsApp/SMS integration

  4. Upload Knowledge Base:

    • Use the form trigger to upload bootcamp handbooks and operational documents
    • Ensure files are in PDF or CSV format

  5. Test Integration:

    • Send test messages via Telegram or WhatsApp
    • Verify responses are contextually appropriate and reference uploaded documents

  6. Activate Workflow: Set the workflow to active status to begin processing messages

Note: This workflow was created from the "rag-starter-template" and has been customized for SkillUp Bootcamp operations.