📢 Important Notice: All subscriber accounts will be removed by 31 July 2026. Please log in using SSO to access HKU SPACE AI Hub.

Codex

OpenAI Codex is an advanced developer ecosystem, terminal command-line tool (@openai/codex), and desktop application designed to act as an autonomous, agentic coding partner.

Visit OpenAI Codex website for full experience

Remarks

While Codex originally launched as a foundational code-generation API, it has evolved into a robust multi-agent environment powered by OpenAI’s frontier models. It operates directly on your local machine, within your IDE, or in sandboxed cloud worktrees to execute end-to-end software engineering tasks.

Key Usages & Workflows

  1. Autonomous Feature Building & Refactoring: You can prompt Codex with a complex goal (e.g., “Migrate this entire module from Express to Fastify”), and it will create isolated worktrees, modify code across multiple files, and run tests until the task is complete.
  2. Terminal & CLI Automation: By running npm i -g @openai/codex, developers use commands like /status and /statusline directly in their terminal to monitor context, review local git diffs, and spin up subagents to solve issues.
  3. Always-On Background Automation: It can be hooked into CI/CD pipelines or repositories to autonomously triage incoming issues, generate comprehensive test suites, and write high-signal Pull Request (PR) design overviews.
  4. Desktop & Browser Integration: Through its desktop application (macOS/Windows) and Chrome extension, it supports “Computer Use,” allowing it to interact with local development tools, documentation, and web dashboards to complete cross-app workflows.

Limitation

  1. Strict Rate Limits & Compute Windows: Because running multi-agent code cycles requires intense reasoning power, OpenAI enforces strict daily or weekly time-based limits (e.g., a 5-hour high-priority active coding quota depending on your ChatGPT Plus/Pro/Business tier).
  2. Context Compaction & Latency: When working inside massive monorepos, Codex must constantly compress and compact repository context. This can introduce processing latency or cause the agent to lose track of deeply nested, obscure codebase dependencies.
  3. Security & Environment Guardrails: To prevent harmful actions on your machine, Codex heavily relies on Model Context Protocol (MCP) approval choices. If security permissions are configured too strictly, the agent will stall; if too loose, it risks executing unintended destructive local terminal commands.
  4. Logical Hallucinations in Edge Cases: While highly proficient at standardized frameworks, Codex can still hallucinate or write syntactically correct but structurally flawed logic when dealing with highly proprietary, un-exported internal APIs or legacy, undocumented codebases.

Step 1: Download the App

Go to the official OpenAI Codex page (openai.com/codex) or the Microsoft Store on Windows. Download the installer compatible with your operating system (Windows or macOS) and run the setup file.

Step 2: Log In with Your Account

Launch the newly installed app. Click Sign in with ChatGPT to log in using your active OpenAI/ChatGPT credentials via a secure browser pop-up. Alternatively, you can input a standalone OpenAI API Key from your developer dashboard.

Step 3: Select a Project Folder

Codex operates relative to your code workspace. Click Choose a Project or drag-and-drop an existing folder or local Git repository directly into the app window.

Step 4: Set Up Your Workspace Mode

Select Local mode in your active project thread. This ensures Codex safely scans your files locally before executing any background workflows or scripts.

Step 5: Send Your First Diagnostic Prompt

Instead of jumping straight into editing complex logic, ask Codex to parse your workspace first. Type the prompt below to verify its structural awareness:
“Please read through this project folder and give me a brief overview of its architecture and file layout.”

Remarks 1: Access Codex issues

To access Codex from a country where it is not officially supported, a VPN is necessary. An alternative solution for overcoming these access issues is Deepsider or Opera Browser.

Remarks 2: Official Support Resources

To learn the fundamental functions of Codex, explore the following resources for more comprehensive information:
OpenAI Help Center: https://help.openai.com/en

01

Introducing the Codex app

02

Codex for (almost) everything

03

It’s time to fly | Codex

04

Introducing Sites in Codex

Visit Deepseek website for full experience

Remarks

DeepSeek is an AI-powered tool designed for deep information retrieval, analysis, and content generation. It is commonly used in areas such as:

  1. Advanced Information Retrieval
    • DeepSeek can process and analyze large datasets to extract relevant insights.
    • It helps users find precise information beyond standard search engines.
  2. Natural Language Processing (NLP) Applications
    • Used for text summarization, sentiment analysis, and question-answering systems.
    • Supports various languages and can generate human-like responses.
  3. AI-Assisted Research and Writing
    • Helps researchers analyze academic papers, generate summaries, and suggest references.
    • Useful for drafting articles, reports, and creative writing.
  4. Code Assistance and Debugging
    • Provides AI-powered code suggestions, optimizations, and bug fixes.
    • Supports multiple programming languages, aiding developers in software development.
  5. Business and Decision-Making Support
    • Analyzes market trends, customer feedback, and financial data for businesses.
    • Assists in generating insights for strategic decision-making.

limitation:

  1. Accuracy and Hallucination Issues
    • AI models can sometimes generate incorrect or misleading information.
    • Requires human verification before relying on outputs.
  2. Limited Real-Time Data Access
    • May not always provide the latest information if it’s not connected to live data sources.
    • Some AI models work with pre-trained datasets, limiting real-time updates.
  3. Context Limitations
    • Struggles with highly nuanced or ambiguous queries.
    • Long conversations may lead to context loss or inconsistencies.
  4. Ethical and Bias Concerns
    • AI models can reflect biases present in training data.
    • Requires careful consideration when used in sensitive applications.
  5. Computational Resource Constraints
    • Running deep learning models requires significant computational power.
    • Latency issues may arise during complex queries or large-scale data analysis.

Github

X

Step 1

Visit the DeepSeek website. Click on “Start Now” and fill in your details.
Verify your email address to activate your account.

Step 2

Log in to your DeepSeek account.
Navigate to the “Chat” section to start using the chatbot.

Step 3

Type your question or request in the chat box. Press “Enter” to send your message. The chatbot will respond with relevant information or assistance.


Writing Assistance: Ask the chatbot to help with writing tasks, such as drafting emails, creating content, or summarizing text.
Translations: Request translations for text in different languages.
Information Retrieval: Ask questions to get quick answers on various topics.

Remarks.

You can view and manage your chat history.Save important conversations for future reference. Customize Your Experience:

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Scroll to Top