AI Coding Tools
-
Agentmemory hits #2 on GitHub trending: 92% token reduction for Claude Code
Agentmemory is an MCP server that gives Claude Code, Codex CLI, Cursor, Gemini CLI, and Hermes persistent memory across sessions. It runs silently in the background, captures what your agent does, compresses it with AI, and injects the relevant context when the next session starts. No manual save, no fiddling with CLAUDE.md. ## The numbers… Continue reading
-
Cline SDK opens the runtime behind 5M coding agent installs
Cline shipped @cline/sdk on May 13. The agent runtime that’s been running inside Cline’s VS Code extension across 5M+ installs is now a standalone TypeScript package. npm i @cline/sdk and you’ve got it. The team migrated their own CLI and Kanban onto the SDK first; VS Code and JetBrains extensions are next in line. What’s… Continue reading
-
OpenAI Codex Anywhere (Mobile Remote Control): scan a QR, drive Codex from your iPhone
OpenAI just put Codex inside the ChatGPT mobile app. Open Codex on your Mac, hit pair, scan the QR with your phone — now your iPhone is a live cockpit for whatever Codex is doing on the host. Approve commands, switch models, eyeball diffs, watch terminal output, scan test results. Files and credentials never leave… Continue reading
-
Superpowers by Jesse Vincent hits 190K GitHub stars, now the 1 Claude Code plugin
Jesse Vincent (obra on GitHub) shipped Superpowers and it’s currently pulling in 1,400+ stars a day. The repo just crossed 190K total. More telling: Anthropic added it to the official Claude Code plugin marketplace, where it’s already the most-installed plugin in the store. What Superpowers actually is It’s an agentic skills framework. Vincent took the… Continue reading
-
Matt Pocock Skills crossed 67K stars — 3,886 added in 24 hours
Matt Pocock published his personal .claude/skills/ directory as a public GitHub repo and watched it eat the trending page. 67K stars total, 3,886 in the last 24 hours alone. The TypeScript educator behind Total TypeScript turned his private Claude Code config into the de facto reference for what production agent skills look like. What’s actually… Continue reading
-
9router pulled 941 GitHub stars in 24 hours routing Claude Code to a free backend
A new GitHub project called 9router picked up 941 stars in a single day. It’s not a model, not an agent. It’s a routing layer that lets you point Claude Code, Codex CLI, Cursor, or Cline at a free backend instead of paying for your own Anthropic or OpenAI key. What 9router actually is Think… Continue reading
-
Entire (by Thomas Dohmke) + Checkpoints: $60M seed bets Git wasn’t built for AI agents
Thomas Dohmke left GitHub in August 2025. Six months later he’s back with Entire — $60M seed at a $300M valuation, the largest dev-tool seed on record. Felicis led, with Madrona, M12, Basis Set, Jerry Yang, and Datadog’s Olivier Pomel piling in. What Entire actually is A new dev platform built around one thesis: Git… Continue reading
-
Anthropic Claude Security goes public beta after 500+ CVEs found in closed preview
Anthropic just moved Claude Security from closed preview to public beta. Claude Enterprise gets it first, Team and Max next. Opus 4.7 reads your entire codebase, traces data flows across files and modules, and tells you exactly where you’re bleeding. What the agent actually does It lives inside Claude Code on the web and runs… Continue reading
-
agentmemory (rohitg00) hits 95.2% on LongMemEval — 27 points above Mem0
Coding agents forget. Close Cursor, reopen tomorrow, re-explain the project. agentmemory fixes that — a TypeScript memory layer climbing GitHub trending fast: +518 stars yesterday, sitting at 3,196. What it actually is A persistent memory server that drops into Claude Code, Cursor, Gemini CLI, OpenCode, or any MCP client. All agents share one memory. It… Continue reading
-
Subquadratic SubQ claims 1,000x less compute at 12M tokens — researchers want receipts
A Miami startup nobody had heard of two weeks ago shipped what it calls the first fully sub-quadratic commercial LLM. SubQ runs a native 12-million-token context window. The numbers Subquadratic put on the page: 50x faster and 50x cheaper than frontier models at 1M tokens, and roughly 1,000x less compute at the full 12M window.… Continue reading
