With AI, Everyone is a Dev
Checking profile…
Fresh AI dev tools
View all new tools →
Claude Duet
Proactive AI Agent by Message
Anthropic's proactive agent you message by text or voice. It briefs you daily, works in your connected apps on its own cloud computer, and asks before it sends, pays, signs, or deletes.
Moshi
Mobile SSH Terminal for AI Agents
A mobile terminal app for running and monitoring coding agents on your own machines over SSH, Mosh, or ET.
Elephant-Goldfish Model
Spec Drafting Plugin for AI Coders
Open-source plugin for AI coding agents that drafts software specs with deep-context agents and reviews them with zero-context subagents.
Rat Stack
TypeScript Full Stack Template
An open-source TypeScript scaffold and reference site for building Effect and Alchemy applications with agent-facing MCP, HTTP, and CLI interfaces.
This Week’s AI Dev Gems
Handpicked tools for the week of Oct 31–Nov 6
Runs a real macOS guest on Apple silicon so coding agents can click, type, and screenshot inside an isolated Mac streamed into your terminal.
Why: Created Nov 3 and already ~2.1k GitHub stars with Daily Launch #11 (score 72 on Nov 5); builders on X are circulating it as a second Mac for Claude Code/Codex that never hijacks the host pointer.
Claude Code and Codex plugin that draws an interactive plain-English map of your codebase - parts, files, web requests, and connections - and flags risks like leaked keys.
Why: Repo created Oct 27 and already ~3.1k stars; X posts this week cite +424 stars in a day and ~2.3k in a week, plus floorplan--v0.6.3 on Nov 2 - the vibe-coded-app map people keep asking for.
DeepSeek's proactive coding agent for one GitHub repo that turns requests into work items and pull requests, shipped as a DeepSeek Harness bundle.
Why: Repo created Oct 31 and already ~20k stars / 2.3k forks; X saw ~2,500 posts mentioning dsh-superagent this week, including Machine Heart's launch note (~1.7k impressions).
Lossless headless coding-agent harness as a single ~7 MB Go binary: four tools, stdin/stdout, and no silent context truncation - it errors if the model hits the limit.
Why: Created Nov 1 with v0.1.0/v0.1.1 the same day; ~610 stars and posts claiming +517 on day one, with builders highlighting the hard 'never silently cut context' rule.
AI Dev News
88+ articles
Ideas from the community
Compare Tools
Ask for a solution, not another list.
“I need a self-hosted browser agent that can work with my existing stack. What should I look at?”
Two models search the catalog
Explore their recommendations and reasoning.
You choose the useful answer
Compare perspectives, then investigate the tools.
What would you let an agent do when a production alert fires?
Investigation, repair and deployment can have different approval rules. Choose the furthest step you would allow without human approval.
Keep learning while you build
Explore developer podcasts and the full AI stack guide between deep dives.