Qodo Brings Organizational Context and Code Review to Codex
Qodo is integrating company knowledge and coding standards into Codex to help developers catch critical cross-repository issues before opening a pull request.
Radwan AbdulhadiEditor and technology writer
Integrating Company Knowledge Into Codex Sessions
Qodo has announced a new integration that brings organizational context and codebase standards directly into Codex. Known for building AI tools for code review and software quality, <a href="https://www.qodo.ai/">Qodo</a> gives coding agents the specific knowledge they need to understand the code they change. Co-founder and CEO Itamar Friedman refers to this capability as the coding agent's best friend. Further details are available from OpenAI News in the original source material.
The platform relies on OpenAI models, including GPT-6 Astra, to gather engineering knowledge, review code, and monitor overall software quality. By introducing a dedicated plugin into the Codex marketplace, the company aims to bridge the gap between local coding tasks and broader enterprise systems.
Catching P0 Bugs Across Large-Scale Repositories
Customer environments can frequently span 10,000 or more repositories, creating scenarios where a localized code change appears correct on its own but breaks a downstream service. To address this challenge, Qodo developed its plugin to give Codex access to complete codebase knowledge, organizational rules, and independent reviews through Qodo's Agentic Toolbox.
With this added context, Codex can account for dependencies spread across multiple repositories, including those hosted by different Git providers or services connected through Kafka. According to Itamar Friedman, this allows the system to look at software as a whole and prevent critical P0 bugs from slipping through.
Automating Code Review While Preserving Human Judgment
Before Codex writes code, it can use the Qodo plugin to pull relevant codebase rules, pull request history, and established patterns. Once Codex completes the changes, Qodo reviews them and returns its findings directly to Codex in the same development session, all before the developer opens a pull request.
Nnenna Ndukwe, Qodo's AI Developer Relations Lead, notes that developers are utilizing this streamlined workflow to build features and address technical debt using GPT-6 models. By automating the handoff between the coding work of Codex and Qodo's independent review, teams can optimize their development cycles while maintaining high standards.
The Yin and Yang of Coding and Review Agents
Itamar Friedman describes coding agents and review agents as yin and yang: one creates code while the other challenges it. Each agentic role requires its own setup and context. The long-term vision involves agents discussing their findings internally and surfacing only unresolved issues for human review.
Under this framework, developers act primarily as planners and reviewers, inspecting discussions, context, and prior decisions alongside the agents. This collaborative dynamic is designed to build trust and increase overall operational efficiency.
Leveraging the Decisions API for Streamlined Rules
Looking ahead, Qodo sees substantial potential in using OpenAI's recently launched Decisions API to determine which specific rules and agent skills apply at each stage of development. Rather than relying on language models for every single decision, the Decisions API could help filter rules more rapidly and economically.
For instance, if a project contains hundreds of rules, the API can quickly identify which subset requires checking before a language model reasons through them. This approach makes rule selection faster and cheaper while maintaining review quality.
Building Owned Software Factories
Qodo's overall methodology combines the advantages of artificial intelligence with traditional software engineering and scripting. Company knowledge guides the coding agents, independent review validates their work, and human expertise resolves remaining questions.
This foundation supports the creation of software factories that customers can assemble and own, uniting coding and review agents with a clear understanding of production software changes. Developers can learn more about related initiatives and startup resources by visiting <a href="https://openai.com/startups">Start building</a> for further details.
Sources
- OpenAI NewsQodo builds context into AI code review with Codex
Continue chronologically
Related entity coverage
Related stories
Proaction boosts sales 60% and saves 75+ hours with Codex
Discover how Proaction uses OpenAI's Codex and GPT models to build, operate, and sell modern fleet management software faster.
Sullivan & Cromwell Develops Agreement Analyzer with OpenAI
Sullivan & Cromwell has collaborated with OpenAI to develop Agreement Analyzer, an AI tool designed to identify issues and review draft agreements.
OpenAI Introduces GPT-6.1 Sol Model
OpenAI has introduced GPT-6.1 Sol, bringing near-Astra intelligence to coding, computer use, and professional workflows at lower token pricing.
Airbnb expands access to GPT-6 Astra for engineering teams
Airbnb is widening access to OpenAI frontier models, including GPT-6 Astra, to help engineering teams solve bugs, design systems, and ship features faster.
