Skip to content

The AI and Human powered ticket system to manage your AI based code generation with tickets

License

Notifications You must be signed in to change notification settings

jmikedupont2/ai-ticket

Folders and files

NameName
Last commit message
Last commit date

Latest commit

author
mike dupont
Oct 20, 2023
d737db4 · Oct 20, 2023
Oct 11, 2023
Oct 19, 2023
Oct 19, 2023
Oct 6, 2023
Oct 20, 2023
Oct 11, 2023
Oct 11, 2023
Oct 20, 2023
Oct 11, 2023
Sep 28, 2023
Oct 11, 2023
Oct 20, 2023
Oct 11, 2023
Oct 15, 2023
Oct 11, 2023
Sep 24, 2023
Oct 11, 2023
Oct 19, 2023
Oct 19, 2023
Oct 6, 2023
Oct 19, 2023
Oct 11, 2023
Oct 11, 2023
Oct 11, 2023
Oct 16, 2023
Sep 29, 2023
Oct 11, 2023
Sep 29, 2023
Oct 16, 2023
Oct 16, 2023

Repository files navigation

AI-Ticket

The AI Ticket system to handle the AI with tickets. Human Powered AI-Ops to Help you with the last mile of your AI code generated system.

Summary: User-Driven Ticket-Based Workflow

Welcome to our innovative user-driven workflow, designed to empower you to interact with our system like a large language model. Here's a quick overview of what we're doing:

1. User as Language Model:

  • You take on the role of a large language model proxy.

2. Step-by-Step Interaction:

  • Each interaction with the system occurs step by step.
  • At each step, you generate a query or response, which becomes a "ticket."

3. Proxy Server:

  • A proxy server facilitates your interactions, creating a ticket for each step.

4. Request Assistance Action:

  • AutoGPT integrates with this workflow using a "Request Assistance" action.
  • This action includes the URL of the GitHub ticket comment linked to your query.

5. User-Controlled Process:

  • You guide the workflow, making decisions, and generating tickets at each step.

6. Ticket-Based Workflow:

  • Each ticket records a step in your interaction, ensuring documentation and continuity.

7. AutoGPT's Role:

  • AutoGPT waits for ticket updates with valid responses generated by you.
  • It retriggers actions based on your responses to move the workflow forward.

8. Potential Stalls:

  • Be mindful of providing timely and appropriate responses to prevent workflow stalls.

This user-driven ticket-based workflow offers flexibility, control, and a unique way to collaborate with our system. Dive in, generate tickets, and explore the possibilities of this interactive and dynamic approach.

Infrastructure

The docker images are organized like this :

  • act_base is the foundation of all actions.
  • poetry_base is contains the poetry magic layer with shared containers.

Importing new agents

we start with the results of the download of the change stats Auto-GPT/stats/filtered_result2.csv

Running

sudo docker-compose build sudo docker-compose up mockopenai sudo docker-compose run auto-gpt-chat-gpt-prompts

Example of verbose build

sudo docker-compose --progress=plain --log-level DEBUG --ansi=never --verbose build ghostcoder

About

The AI and Human powered ticket system to manage your AI based code generation with tickets

Topics

Resources

License

Stars

Watchers

Forks