AIToday
Large Language ModelsAI Coding AssistantsOpenAI BlogPublished: Nov 3, 2026, 06:00 JST

Qodo's Codex plugin catches P0 bugs across repos

Send on LINE
Qodo's Codex plugin catches P0 bugs across repos

3 Key Points

  1. What happened

    Qodo released a Codex marketplace plugin that feeds Codex codebase rules, pull request history, and independent review via Qodo's Agentic Toolbox, aimed at catching P0 bugs across 10,000 or more repositories before a pull request opens.

  2. Why it matters

    With that context, Codex can account for dependencies across repositories, so a local change that looks correct on its own does not silently break a downstream service.

WHO IT HITSDevelopers working in large multi-repository codebases are the immediate users, since the plugin automates the handoff between Codex writing code and Qodo reviewing it before a pull request. Engineering teams running coding agents across many services may see fewer defects that only surface after deployment.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Qodo's plugin addresses a specific failure mode Itamar Friedman describes: in customer environments spanning 10,000 or more repositories, a local code change can break a downstream service even when it looks correct on its own. By giving Codex access to codebase knowledge, organizational rules, and independent review through Qodo's Agentic Toolbox, the plugin lets the coding agent consider dependencies across repositories, including those hosted by different Git providers or connected through Kafka.

The workflow places review before the pull request. Before Codex writes code, it can retrieve relevant codebase rules and pull request history; after the changes, Qodo returns its findings into the same session. Nnenna Ndukwe says developers are using this to build features and tackle technical debt, automating the handoff between coding and independent review.

Friedman frames coding and review agents as complementary roles, with a vision of agents discussing findings and surfacing only what they don't accept for human review. He also points to OpenAI's recently launched Decisions API as a way to choose which rules apply at each stage, which he says could make selection faster and cheaper while improving review quality.

FAQ
When does the review happen?
Qodo reviews Codex's changes and returns findings directly to Codex in the same session, before the developer opens a pull request.
What is the Decisions API used for?
Itamar Friedman sees it selecting which of 200 rules need checking at each development stage, making selection faster and cheaper rather than relying on a language model for every decision.
Which OpenAI models does Qodo use?
Qodo uses OpenAI models including GPT‑6 Astra to gather engineering knowledge, review code, and monitor software quality.
Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleTUKURO launches AITUKAO from ¥1,500/month with ChatGPT, Claude, Gemini