Listen to this Post
GitHub is taking a major leap in AI-assisted coding with the public preview of Claude by Anthropic and OpenAI Codex as coding agents. These AI-powered assistants are now available for Copilot Pro+ and Copilot Enterprise subscribers, enabling developers to automate coding tasks directly from GitHub, GitHub Mobile, and VS Code. By integrating seamlessly with issues, pull requests, and agent sessions, these tools promise to enhance productivity, reduce repetitive work, and offer intelligent code suggestions across multiple platforms—all without requiring additional subscriptions.
Public Preview of Claude and Codex
GitHub users can now start agent sessions and assign tasks to Claude and Codex directly from repositories where Copilot is enabled. These coding agents work in real-time and provide draft pull requests, making collaboration smoother and coding more efficient. Each agent session consumes a single premium request during this public preview period.
For Copilot Pro+ subscribers, enabling these agents is simple: navigate to Copilot coding agent settings, choose the repositories they can access, and toggle Claude, Codex, or both on. Copilot Enterprise subscribers require admin-level configuration, which includes enabling agents at both enterprise and organization levels through “Enterprise AI Controls” and the Copilot tab in organization settings.
How Developers Can Use the Agents
Creating sessions with Claude or Codex is straightforward. On GitHub.com, users can access the Agents tab in enabled repositories or initiate tasks via the agents dropdown. The GitHub Mobile app also supports session creation from the home or repository view, offering a seamless on-the-go experience. Developers simply type their request, select the preferred agent using the Copilot icon, and submit it. Sessions can be monitored in real time, and completed work can be reviewed later.
Assigning Agents to Issues and Pull Requests
GitHub allows direct assignment of issues to Claude, Codex, Copilot, or any combination, automatically generating draft pull requests. These agents are capable of following up on feedback, iterating on code, and collaborating on open pull requests. Developers can leave comments or request changes using @claude, @codex, or @copilot, and session logs provide a complete overview of the agent’s activity.
Using Claude and Codex in VS Code
VS Code users (version 1.109 or later) can access agent sessions by clicking the chat icon or via the Command Palette. Sessions are available in three types:
Local: Interactive help on the spot
Cloud: Autonomous tasks running on GitHub
Background: Asynchronous tasks locally (Copilot only)
These options allow developers to tailor the AI assistance to their workflow, whether for fast suggestions or long-running tasks.
Additional Resources
GitHub provides multiple ways to get the most out of these agents:
Full blog post
detailing agent features
Third-party agent documentation for advanced usage
Discussion posts for feedback and user experiences
Mobile changelog for creating and managing agent sessions
What Undercode Says:
Impact on Developer Productivity
Claude and Codex represent a paradigm shift in software development. By automating repetitive tasks and generating draft pull requests, these agents reduce the time developers spend on boilerplate code and minor bug fixes. Teams can now focus on creative problem-solving, higher-level architecture, and complex feature implementation.
Integration Across Platforms
GitHub’s decision to make the agents accessible via web, mobile, and VS Code reflects an understanding of modern developer workflows. Whether coding on the go or deep in an IDE, users can maintain continuity and flexibility without switching tools or platforms.
Ease of Adoption
The lack of additional subscriptions is a significant advantage, especially for smaller teams or solo developers. The ability to toggle agents per repository or enterprise ensures granular control, allowing organizations to adopt AI coding assistance at their own pace.
Collaboration and Iteration
With the ability to assign issues or pull requests to AI agents and iterate based on feedback, development cycles can become more agile. AI-assisted code reviews and draft PRs accelerate the workflow while maintaining human oversight for quality control.
Long-Term Implications
Claude and Codex could redefine coding norms over the next few years. As these agents evolve, they may handle more complex tasks autonomously, potentially shifting the role of developers toward supervisory and creative functions.
Challenges and Considerations
Despite the benefits, reliance on AI agents carries risks such as over-dependence, subtle coding errors, and intellectual property concerns. Organizations must implement review processes and maintain clear accountability for AI-generated code.
Future Outlook
The public preview phase offers critical feedback that will shape agent capabilities. Developers who engage early gain a competitive edge in integrating AI into workflows and experimenting with new coding paradigms.
Community and Ecosystem Growth
Third-party agent support hints at an expanding ecosystem. Over time, community contributions and plug-ins may expand the agents’ functionality far beyond their current capabilities, fostering innovation across industries.
Strategic Advantage for GitHub
By integrating Claude and Codex into Copilot subscriptions, GitHub strengthens its position as the go-to platform for AI-assisted development, competing directly with other AI coding tools and cementing loyalty among enterprise clients.
🔍 Fact Checker Results
✅ Claude is developed by Anthropic; Codex is by OpenAI.
✅ Both agents are accessible via GitHub web, mobile, and VS Code.
❌ No extra subscription is required beyond Copilot Pro+ or Enterprise.
📊 Prediction
Claude and Codex will likely see rapid adoption among enterprise teams and open-source contributors. Within 12 months, we can expect AI-generated pull requests to become a standard part of many repositories, reducing mundane coding tasks by up to 40%. This may also trigger competition, pushing other AI coding tools to enhance integration and autonomy features, accelerating the evolution of AI-assisted software development.
If you want, I can also create a visual workflow diagram showing how Claude and Codex integrate across GitHub web, mobile, and VS Code, which could make this article even more engaging. Do you want me to do that?
🕵️📝✔️Let’s dive deep and fact‑check.
References:
Reported By: github.blog
Extra Source Hub (Possible Sources for article):
https://www.discord.com
Wikipedia
OpenAi & Undercode AI
Image Source:
Unsplash
Undercode AI DI v2
Bing
🔐JOIN OUR CYBER WORLD [ CVE News • HackMonitor • UndercodeNews ]
📢 Follow UndercodeNews & Stay Tuned:
𝕏 formerly Twitter 🐦 | @ Threads | 🔗 Linkedin | 🦋BlueSky | 🐘Mastodon
