Top Builders

Explore the top contributors showcasing the highest number of app submissions within our community.

OpenAI Codex

OpenAI Codex is an artificial intelligence model developed by OpenAI. It parses natural language and generates code in response. It is used to power GitHub Copilot, a programming autocompletion tool. Codex is a descendant of OpenAI's GPT-3 model, fine-tuned for use in programming applications. OpenAI has released an API for Codex in closed beta. Based on GPT-3, a neural network trained on text, Codex has additionally been trained on 159 gigabytes of Python code from 54 million GitHub repositories. You can find more information here https://openai.com/blog/openai-codex/

General
Relese dateAugust 31, 2021
AuthorOpenAI
Repository-
TypeAutoregressive, Transformer, Language model

Start building with Codex

We have collected the best Codex libraries and resources to help you get started to build with Codex today. To see what others are building with Codex, check out the community built Codex Use Cases and Applications.

Boilerplates

Kickstart your development with a Codex based boilerplate. Boilerplates is a great way to headstart when building your next project with Codex.

  • Codex Boilerplate Create a function just by typing what it should do, with help of OpenAI Codex.

Libraries

A curated list of libraries and technologies to help you build great projects with Codex.

Edit on GitHub

OpenAI Codex AI technology Hackathon projects

Discover innovative solutions crafted with OpenAI Codex AI technology, developed by our community members during our engaging hackathons.

AetherDev Pro

AetherDev Pro

AetherDev Pro is an advanced, production-ready multi-agent software development platform and interactive IDE designed for automated software engineering workflows. Built on a Flask backend and a premium glassmorphic HTML/CSS/JS frontend, the platform integrates Microsoft Monaco Editor (the core engine of VS Code) to allow developers to view, edit, and save generated files in real-time. Key Features & Agent Workflow: 1. **Multi-Model Agent Teams**: Users can customize their AI engineering team by routing specific LLMs (e.g. Google Gemini 1.5 Pro, Llama 3.3 70B, GPT-4o) to specialized roles: - **Planner Agent**: Analyzes prompts and outputs structural design layouts and DAGs. - **Engineer Agent**: Automatically implements code for planned files. - **Reviewer Agent**: Evaluates syntax, error handling, and logical correctness, requesting iterative improvements. - **Tester Agent**: Autonomously writes test suites using python's unittest framework. - **Documenter Agent**: Generates comprehensive README files and code documentation. 2. **Self-Healing Code Compilation (TDD Loop)**: AetherDev Pro executes generated test suites in a secure local sandbox subprocess. If any test fails, the error traceback is dynamically parsed and fed back to the Engineer agent with instructions to repair the codebase. This loop repeats autonomously until all tests pass, ensuring that the final output is verified and functional. 3. **Stateless Persistence (SQLite)**: All sessions, file trees, source contents, run records, and terminal logs are persisted in a local SQLite database. This keeps the application robust, resilient to server restarts (such as on cloud platforms like Render), and allows users to resume past projects seamlessly.

Periop Band: Anesthesia Safety Agents

Periop Band: Anesthesia Safety Agents

Periop Band is a doctor-in-the-loop multi-agent workbench for high-stakes perioperative anesthesia coordination. It ingests synthetic pre-op notes, ECG and lab summaries, imaging text, and intraoperative event updates, then produces structured, reviewable drafts for anesthesiologists rather than autonomous clinical decisions. The Band-ready collaboration layer defines specialized agents for perioperative intake, ECG and lab risk review, safety review, and postoperative surveillance. These agents pass structured context, hand off tasks, create an auditable collaboration trace, and keep escalation visible for the clinician. The demo includes a React and FastAPI interface, OpenAI Agents SDK backend workflow, safety guardrails, local eval cases, Markdown transcript export, and live Band configuration hooks.

HazardMind AI

HazardMind AI

HazardMind AI is a multi-agent disaster response system built for the Band of Agents Hackathon. When a disaster is reported, four specialized AI agents collaborate through Band to deliver a complete situational analysis in under 60 seconds. Agent 1 — Satellite Agent fetches real Sentinel-1 SAR and Sentinel-2 optical imagery from Copernicus, automatically selecting the optimal satellite based on cloud cover and disaster type. It processes multi-band imagery using NDWI, NDVI, and SAR indices to produce georeferenced classification maps and vectorized risk zones. Agent 2 — Hazard Detection Agent cross-validates satellite findings with GDACS and USGS real-time disaster data to confirm risk zones and severity levels across flood, earthquake, and landslide scenarios. Agent 3 — Impact Assessment Agent overlays WorldPop population density data with OSM infrastructure to calculate affected populations, hospitals at risk, blocked roads, and optimal evacuation routes via OpenRouteService. Agent 4 — Executive Report Agent compiles all findings into an interactive MapLibre dashboard with satellite overlays, risk zone layers, and a downloadable PDF report with AI-generated executive summary. All four agents communicate through Band — confirming findings, escalating critical discoveries, and maintaining a visible discussion log that provides full auditability of every decision made during the response. Built for NDMA, UN agencies, and disaster response organizations worldwide.

Wenties Sentinel Desk

Wenties Sentinel Desk

Wenties Sentinel Desk is a practical enterprise multi-agent workflow for incident awareness, operational visibility, and controlled decision-making. Many small and regulated organisations receive alerts, reports, access issues, support requests, and operational risks, but the decision process is often scattered across messages, screenshots, and undocumented judgment calls. This project uses Band as the coordination layer between specialised agents. A Triage Agent classifies the incoming event and extracts key facts. A Risk Agent evaluates severity, business impact, and escalation need. An Operations Agent recommends practical next steps. A Review Agent checks the recommendation for clarity, missing context, and audit readiness before a human manager approves the final action. The goal is not to replace humans. The goal is to make complex operational decisions more structured, traceable, and reviewable. The demo shows agents sharing context through Band, handing off work, challenging weak recommendations, and producing a final decision record with status, rationale, timeline, and audit trail. The prototype uses sanitized demo data only. It is designed for enterprise workflows where cyber resilience, compliance awareness, internal coordination, and human-reviewed escalation matter.

BunoBagera AI

BunoBagera AI

I built Bunobagera AI because I wanted to bridge the gap between writing code and shipping production-ready software. This isn't just a hackathon prototype—it is a robust, production-grade application engineered for real-world reliability. The frontend is built on Next.js for a lightning-fast, seamless UI, while the backend leverages Strapi as a powerful, flexible headless CMS. For data storage, I integrated Supabase to handle database management reliably at scale. At its core, the application uses the Mistral LLM to effectively analyze codebases for security, efficiency, and relevance. When you connect a repository, Bunobagera scans your code and enters an interactive chat interface to discuss fixes. The best part is the automation: if you approve the suggestions, it automatically creates a new branch and opens a Pull Request with the fixes. Beyond the code, it reviews your GitHub profile, providing a strategic summary and a personalized roadmap to help you grow your developer brand and stand out to recruiters.

SovereignQA: 7-Agent Self-Healing DevOps Mesh

SovereignQA: 7-Agent Self-Healing DevOps Mesh

SovereignQA is an autonomous, state-driven multi-agent DevOps framework designed to replace fragile, linear CI/CD pipelines with a self-healing QA council. Built entirely on top of the stateful Band.ai protocol, the platform creates a decentralized network where specialized AI agents collaborate asynchronously using an isolated data ledger (Band Room) as their absolute source of truth. The operational lifecycle is triggered natively via GitHub webhooks upon a code push or Pull Request activation. Instead of step-by-step sequencing, the system uses non-linear state orchestration split across three discrete validation rings: 1. Ingestion & Static Verification: Micro-agents execute static syntax diagnostics (Linter Agent), map code paths against security risk profiles (SecOps Auditor for OWASP vulnerabilities), and validate type-hint definitions (Schema Watchdog). 2. Dynamic Runtime Execution: A dedicated Pytest Assert Engine compiles structural assertions, executing code inside an ephemeral, sandboxed Docker container to safely monitor runtime exceptions, while a UI Vision Layout Agent reviews DOM element alignment. 3. Autonomous Remediation & Feedback: If execution fails, a Self-Heal Core agent intercepts command-line tracebacks from the ledger, computes programmatic fixes, patches source files, and loops the state machine back to re-trigger testing. Once cleared, a GitHub Notifier agent posts a comprehensive markdown dashboard and copy-pasteable Git diff right into the developer's pull request. SovereignQA addresses real-world enterprise constraints by introducing an asynchronous message queue (Redis/RabbitMQ) to flatten transaction spikes, sandboxed containerization for secure code processing, and loop kill-switches to protect API token budgets. This ensures a robust, secure, and highly scalable platform.

Codex