SECAD

Summary

This repo contains the containers for running SECAD. SECAD is an agentic, AI-powered security workflow augmentation application. See the main project here - https://github.com/gideonaina/secad

For more details on design and use case, read this article: https://medium.com/@gideonaina/workflow-augmentation-with-multi-agent-ai-system-3c3223c948dc

Setup

Pre-requisite (Required)

• Clone this repo: git clone git@github.com:gideonaina/secad-lite.git
• Install Docker (recommend v27). It comes with docker compose (minimum needed is v2, v1 version will not work with this lab).
  • Mac: https://docs.docker.com/desktop/setup/install/mac-install/
  • Windows: https://docs.docker.com/desktop/setup/install/windows-install/
  • Run docker --version to confirm the version of Docker.
  • Run docker compose version to confirm the version of docker compose.

Pre-requisite (Optional)

• Install pgAdmin: Used to inspect the content of the vector database

  • Mac: https://www.pgadmin.org/download/pgadmin-4-macos/
  • Windows: https://www.pgadmin.org/download/pgadmin-4-windows/

• Just: Makes it easier to run the necessary commands as recipes (See Justfile in the repo for details)

  • Mac: run brew install just
  • Window:
    • Install Scoop or Chocolatey

    Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser
    
    Invoke-RestMethod -Uri https://get.scoop.sh | Invoke-Expression
    

    • Once installed, run scoop install just

• Install VS Code: https://code.visualstudio.com/download

Running SECAD

• Go to https://platform.openai.com/settings/organization/api-keys, register/log in to your account, and create an API key. Load some money into your account ($10 is more than enough).
• Create a .env at the root of this project and copy the content of the .env.sample file into it. Provide the OPENAI_API_KEY value. Leave the rest as default.
• Run the just command just start or copy the command into the recipe and run it from your command line docker compose -f secad.docker-compose.yml up
• SECAD should be running on localhost on port 8501. i.e. localhost:8501