Jean
Jean, an open-source development environment for AI-assisted coding, is available under the Apache 2.0 license, free forever with no paid tiers or feature gates. It enables running multiple AI agents …
Jean, an open-source development environment for AI-assisted coding, is available under the Apache 2.0 license, free forever with no paid tiers or feature gates. It enables running multiple AI agents …
Chaitanya Giri's Munder Difflin v0.4.5 fixes inaccurate cost reports, broken semantic memory on Apple Silicon, and unreliable communication between AI workers in the local-first agent harness that wra…
A census of 12 coding-agent command-line interfaces found that file-write permission defaults in headless mode split three ways: 7 ask or propose until a flag is passed, 2 refuse to write until flagge…
Moonshot released the weights for its 2.8-trillion-parameter Kimi K3 model on July 27, 2026, 11 days after launching the model across Kimi, Kimi Work, Kimi Code, and the Kimi API, extending CEO Yang Z…
Cosyncing, a new open-source tool, lets users synchronize and control coding agents such as DeepSeek Harness and Claude Code from a phone or desktop via a local broker, with no hosted service. The bro…
A documentation census of 17 coding agents and harnesses, published on August 17, 2026, found that all 17 transmit code off the local machine in at least one common configuration, and 8 of 17 rows are…
A developer has documented a workaround for running Kimi Code's K3 model inside Anthropic's Claude Code terminal tool, addressing configuration pitfalls including a misleading model ID and authenticat…
Maestro, a new desktop app for Windows, Linux, and macOS, lets developers run Claude Code, Codex, Cursor, OpenCode, and Kimi Code side by side in separate git worktrees, with a rail showing who's work…
Kimi K3, an AI coding model from Moonshot AI, built a playable first-person 3D browser game from a single prompt, but the game had bugs that only surfaced during human play, such as a drone clipping t…
Kota, an open-source tool that enables AI agent CLIs to collaborate in a shared workspace, was released by its creator after months of personal use. It supports Codex, Claude Code, Pi, OpenCode, Gemin…
A developer reports that the gap between AI models and their agentic performance is often determined by the harness—the system around the model—rather than the model itself. Evidence from Moonshot's K…
Composio, an AI-agent infrastructure company, reported that Moonshot AI's Kimi K3 consumed nearly six times as many tokens in Claude Code (340,000 median) as in Kimi Code (61,000 median) while complet…
Moonshot AI's Kimi Code, a coding agent, is vulnerable to a FetchURL SSRF protection bypass in versions before 0.27.0, allowing attackers to reach internal network services via DNS-resolving hostnames…
A new workflow using Moonshot AI's Kimi K3 coding agent and AI video tools lets a single person build cinematic scroll-driven websites for roughly $1, replacing what previously required a designer, vi…
Moonshot's Kimi Code, a TypeScript rewrite of its Python-based Kimi CLI, opens the entire coding-agent stack under an MIT license, shipping tuned for the open-weight K2.7-Code model on Hugging Face un…
Vercel announced that its Vercel Plugin is now available in the Kimi Code CLI, enabling the AI coding assistant to access Vercel platform knowledge for Next.js, AI SDK, Vercel Functions, and more. Use…
Pi, a minimal terminal-based coding agent designed for maximum customizability, has emerged as the core infrastructure powering the viral OpenClaw AI assistant tool, offering developers a hackable fou…
Nvidia's GEAR lab, in collaboration with Carnegie Mellon University and UC Berkeley, developed the ENPIRE framework that enables AI coding agents to autonomously train fleets of physical robots on pre…
Nvidia, Carnegie Mellon University, and UC Berkeley researchers developed ENPIRE, a system where AI coding agents autonomously train robots in dexterous grasping tasks. Using a fleet of eight robots t…
Moonshot AI released and open-sourced Kimi-K2.7-Code, a coding model that improves agent performance by up to 31.5% on benchmarks and reduces reasoning overhead by 30% compared to K2.6. The model is a…