Coding Agents

Reports say AI coding agents pushed 13,000 internal images to public GitHub repositories, exposing billing records and raising urgent controls questions.

AWS shows developers how to pair OpenCode with open-weight models in Amazon Bedrock, bringing private, flexible, pay-per-use AI coding agents to AWS.

Unity’s plugins for Claude Code and OpenAI Codex add maintained Unity 6 skills, aiming to reduce outdated guidance in AI-assisted game development.

AWS and Hugging Face are showing how six open-source skills help coding agents deploy models on Amazon SageMaker AI with safer, repeatable steps.

OpenAI says coding agents are changing AI research, but its early evidence leaves key questions about scale, measurement, and independent validation open.

Tests of Claude Code and Codex found major errors in time estimates and self-evaluation, raising oversight concerns for long-running AI coding tasks.

Researchers found AI coding agents installing packages tied to abandoned documentation references, exposing a supply-chain risk inside corporate networks.

Block’s free Goose agent gives developers a local alternative to Claude Code, trading cloud convenience and model quality for privacy, control, and lower cost.

Block’s free Goose agent offers local coding automation without subscriptions, giving developers a privacy-focused alternative to Claude Code’s paid limits.

NVIDIA’s ALCHEMI Toolkit connects AI coding agents with GPU-accelerated materials simulation, while benchmarks show validation remains essential.

Meta’s beta Muse Code brings parallel AI agents to large codebases, challenging Codex and Claude Code with a potentially lower-cost workflow.

Meta has introduced its first AI coding agent, putting the company into direct competition with Anthropic and OpenAI as coding tools expand for developers.

An OpenAI-backed report says coding agents can speed research software upgrades dramatically, but experts still must verify scientific correctness.

Poolside released Laguna S 2.1, an open-weight coding model that it says rivals far larger systems by improving persistence and verification.

OpenAI’s Codex now encrypts internal agent handoffs, limiting developer visibility into delegation and raising reliability questions for AI coding workflows.

Nvidia CEO Jensen Huang says more software work is moving from Python coding to AI agents, underscoring a broader shift in how teams build AI products.

OpenAI outlined Codex sandboxing, approvals, network policies, and telemetry for secure coding-agent deployment.