STORY · VERKTOY_
Anthropic, OpenAI and Nous Research launch agent tools for code iteration
Anthropic rolled out computer use in Claude Code with a closed verification loop for Pro/Max users, OpenAI launched a Codex plugin for Claude Code for cross-agent composition, and Nous Research's Hermes Agent achieved rapid adoption. The debate centers on open source versus proprietary agent infrastructure, with focus on tools, orchestration, and feedback loops.
WHY IT MATTERS
This signals that AI agents' ability to iterate on their own code and collaborate across systems is becoming critical infrastructure – it shifts focus from model capability alone to how agents are coordinated and verification loops function.
SOURCES
MACHINE-GENERATED SUMMARY This summary is written by machine from the sources below. We sort and explain — but we are a way into the field, not the final word. Check the source when something matters to you.