DOCSEARCHbyAlgolia
Submit Docs

AI Agents A-Z

repository·main·Indexed 26 days ago

https://github.com/gyoridavid/ai_agents_az

A collection of n8n workflow templates and specialized MCP/REST servers for building AI agents. Use cases include research, lead generation, and automated content creation for YouTube, TikTok, and Shopify. The repository provides JSON workflows for integrating services such as Together AI, Fal.ai, MiniMax Hailuo 2, Bytedance Seedance, and Postiz, as well as deployment guides for Modal.com using FLUX.1 Kontext [dev] and Wan 2.2.

Tokens
10.4K
Snippets
20
Records
99
Agent score
87%

What's inside ai_agents_az

  1. Overview of the Scary story TikTok videos workflow

    main
    This workflow automates the creation of TikTok scary story videos. It uses Together AI for image generation, the AI Agents No-Code Tools media server for processing, and Reddit-inspired stories. Once configured, the workflow can be triggered by filling out a form with background music and a voice sample, which then automatically generates a video with a specific art style and story instructions across multiple scenes.

  2. Create short motivational stories for YouTube and TikTok

    main
    This automation builds an AI short video maker factory that generates motivational stories and publishes them to YouTube. It can be configured to publish to TikTok only and can be run locally using free tools. The workflow uses n8n for orchestration and integrates with services like Together AI for model generation.

  3. Access n8n templates for AI Agents A-Z

    main
    This repository contains n8n workflow templates corresponding to the episodes of the AI Agents A-Z YouTube series. You can find specific automation workflows for various AI agent use cases (e.g., research agents, social media automation, video generation) organized by episode number in the episode_X directories.

  4. Create AI videos with MiniMax Hailuo 2 and n8n

    main
    This project provides a method to automate AI video generation using MiniMax's Hailuo 02 (2nd generation video AI model) integrated with n8n. You can use the provided n8n workflow as a standalone form-based tool or transform it into a subworkflow to build a fully automated AI video factory.

  5. Retrieve the Instagram Business Account ID

    main

    To configure the Instagram node in n8n, you need your Instagram Business Account ID. You can retrieve it using the Graph API Explorer:

    1. In the Graph API Explorer, click on the ID of your Facebook page.
    2. Append ?fields=instagram_business_account to the end of the ID in the graph path.
    3. Click Submit to see the associated Instagram business account details.

  6. Run the AI Agents No-Code Tools server locally

    main

    To run the server on your local machine, ensure you have Docker installed. Use the standard image for CPU-only execution or the -cuda tagged image if you have an NVIDIA GPU with the Cuda Toolkit installed.

    Hardware Requirements:

    • At least 4 vCPUs.
    • For chatterbox TTS: at least 8GB RAM.
    • For kokoro TTS: at least 6GB RAM.

  7. Setup requirements for UGC video workflow

    main

    To run the UGC video automation workflow, you will need access to the following services and API keys:

    • n8n: The orchestration platform for the workflow.
    • Fal.ai: Required for model access via API keys.
    • ElevenLabs: For audio generation.
    • Seedance & Latentsync: For video generation and lip-syncing components.

  8. Run the AI Agents No-Code Tools server with NVIDIA GPU support

    main

    If you have an NVIDIA GPU and the CUDA Toolkit installed, you can run the server with GPU acceleration using the -cuda tagged image. This requires passing the --gpus=all flag and setting the appropriate NVIDIA environment variables.

    docker run --rm --gpus=all -e NVIDIA_VISIBLE_DEVICES=all -e NVIDIA_DRIVER_CAPABILITIES=all -p 8000:8000 -it gyoridavid/ai-agents-no-code-tools:latest-cuda

  9. Create free explainer videos with n8n and Z-Image

    main

    You can automate the creation of explainer videos (including scripts, voiceovers, and visuals) using an n8n workflow combined with a Modal server. This setup allows you to use Kokoro and Z-Image Turbo for free via a Modal account.

    To implement this, you need:

    1. A Modal server to handle the video creation logic.
    2. An n8n workflow to orchestrate the process.

    Required Resources:

    • Modal server: modal_explainer_videos.py
    • n8n workflow: n8n_workflow_explainer_video_creator.json
    • Z-Image-Turbo: Available on Hugging Face.
    • Modal: Used for hosting the server/GPU resources.

  10. Create LinkedIn posts with human-in-the-loop using n8n

    main

    This workflow automates the creation of LinkedIn posts using an n8n AI agent with a human-in-the-loop approval step.

    Important Note on RSS Feeds: Due to traffic limitations from Venturebeat (as of April 2025), the workflow has been updated to use an alternative RSS feed. If you attempt to use the original Venturebeat RSS feed, you may need to use a VPN or a proxy to avoid traffic restrictions.

  12. Create the `plans` database table

    main

    In the Supabase Table Editor, create a new table named plans with the following four fields:

    1. id (int8): Primary key, use default settings.
    2. period_type (text): Used to index different periods. Allowed values: quarter, month, or week.
    3. period_index (text): Stores the specific period value (e.g., 2025-Q2).
    4. plan_value (text): The actual value of the plan for the given period.