Firecrawl, the open-source TypeScript web-scraping and search API for AI agents, gained 8,099 GitHub stars, bringing its total to 171,000 stars as of its last update in August 2026. The AGPL-3.0-licen…
A new open-source project, agent-dev-team, gives AI coding tools a tiered engineering structure with 21 role agents across four seniority tiers, 26 portable workflow skills, and a structured escalatio…
Zvec-ai released zvec-grep (zg), an open-source, local-first hybrid search engine that unifies ripgrep-style regex matching, BM25, and local vector search in a single command-line tool. Built on Aliba…
Void Linux maintainer Andrea Brancaleoni orphaned more than 100 packages he maintained after a dispute over the project's AI usage policy, which requires all AI tool usage to be disclosed in contribut…
Vals.ai's new International Olympiad in Informatics benchmark found that GPT-6 Astra solved every problem across all three years tested, while GPT-5.6 Sol scored 91.17%, Claude Fable 5.1 scored 90.78%…
Novaya-AI released Novgraph, a persistent knowledge graph engine for codebases that records why each change was made, written back by agents as they work. Novgraph syncs its dashboard with commits and…
A new cross-platform desktop app called Maestro runs multiple CLI coding agents — including Claude Code, Codex, Cursor, OpenCode, Kimi Code, and Grok Build — in parallel, each in an isolated git-workt…
Codex can run against a local LM Studio server hosting the 27-billion-parameter gemma-3-27b-it model, but a blocking HTTP proxy test found the tool still contacts four external destinations — chatgpt.…
Orca launched as an open-source agent development environment (ADE) that runs coding agents such as Claude Code, Codex, OpenCode, Grok, and Gemini in parallel across isolated workspaces, with a compan…
A developer evaluated tt-a1i/archify, an open-source tool that compiles AI-generated architecture diagrams through a typed JSON intermediate representation with schema, layout, and routing validation …
Neuro Engine, an MCP server that analyzes a codebase's skeleton so AI coding agents emit compact change blueprints instead of full-file rewrites, claims token savings of up to 96% on large surgical ed…
A developer published quota-tavily, an OpenCode plugin that answers the /quota-tavily command with an instant Tavily API usage report without consuming an LLM turn. The plugin fetches api.tavily.com/u…
Claude Code reads CLAUDE.md while Codex, Cursor, GitHub Copilot, OpenCode, and Amp read AGENTS.md, so developers running multiple agentic harnesses must either maintain two instruction files or link t…
JetBrains's SkillsBench benchmark found that RTK (Rust Token Killer), a terminal-output compression tool with over 79,000 GitHub stars, produced no consistent cost savings for AI coding agents, with D…
A developer guide details running local LLMs with OpenCode, Ollama, and Docker Sandboxes (sbx) on an Apple MacBook Pro M5 with 48GB of memory, pulling Qwen 3.8 27B mxfp8 (32GB) and Gemma 4 31B mxfp8 (…
Developer Sergi Corruchaga released D-Engine v0.2.2, an open-source coding harness that replaces the agentic loop with a deterministic two-call pipeline, cutting token usage by up to 42× versus DeepSe…
Researchers Li, Zhang, Hou and coauthors published an arXiv supply-chain study of AI agent harnesses including Claude Code, Codex CLI, OpenCode, OpenClaw, Hermes, OpenHarness and WorkBuddy, describing…
A team led by @oh_an_opinion built an internal coding benchmark from its own merged pull requests and found that, depending on task complexity, cheaper and faster agent setups matched frontier Anthrop…
A developer has documented a configuration workaround for OpenCode that prevents accidental session loss when users press Ctrl+C out of terminal habit. By remapping the app_exit keybinding in the Open…
DeepSeek released DeepSeek V4.1 Flash, a 552B-parameter MoE model built on a new Causal-Encoder-Decoder architecture with 8B activated parameters on the input side and 16B on the output side, availabl…