gxCaesar published open-research-skills, a GitHub repo with ten independently installable agent skills covering scientific figures, conference and journal manuscripts, funding proposals, method development, cold review, and artifact release. Skills install in two commands as a Claude Code plugin or via script for Codex and Antigravity CLI. It gives researchers ready-made agent workflows across the full research pipeline.
A store selling physical hardware for coding-agent users, including dictation pads, macro pads, a traffic light indicator for agent status, and foot pedals. It aims to improve the ergonomics of interacting with AI coding agents by providing dedicated input devices. The store has had 28 visitors and no sales yet.
A store selling physical hardware for coding-agent users, including dictation pads, macro pads, a traffic light indicator for agent status, and foot pedals. It aims to improve the ergonomics of interacting with AI coding agents by providing dedicated input devices. The store has had 28 visitors and no sales yet.
OpenAI released Codex 0.146.0, introducing session management features like naming, pinning, and forking threads, plus support for Agent Plugins manifests and additional plugin marketplaces. The update also enables standalone web search for compatible custom models.
OpenAI released Codex 0.146.0, introducing session management features like naming, pinning, and forking threads, plus support for Agent Plugins manifests and additional plugin marketplaces. The update also enables standalone web search for compatible custom models.
Aieracard is a command-line tool that reads local usage data from Cursor, Claude Code, and Codex to generate a shareable card showing your token consumption. It prioritizes privacy by showing the exact payload before upload and ensuring the schema cannot contain prompts, code, or file paths. It solves the problem of quantifying and comparing AI coding tool usage for developers who want to showcase their experience.
Aieracard is a command-line tool that reads local usage data from Cursor, Claude Code, and Codex to generate a shareable card showing your token consumption. It prioritizes privacy by showing the exact payload before upload and ensuring the schema cannot contain prompts, code, or file paths. It solves the problem of quantifying and comparing AI coding tool usage for developers who want to showcase their experience.
Loadout is a macOS app that reads configuration files from Claude Code, Codex, and opencode to display all skills, slash commands, and MCP servers in one list. It also analyzes session logs to show usage counts and last-used timestamps for each item, helping users understand what their AI assistants actually load and use.
Loadout is a macOS app that reads configuration files from Claude Code, Codex, and opencode to display all skills, slash commands, and MCP servers in one list. It also analyzes session logs to show usage counts and last-used timestamps for each item, helping users understand what their AI assistants actually load and use.
TokenNations is a leaderboard ranking countries by aggregate AI coding tool token usage instead of individuals. A CLI reads local usage logs from tools like Claude Code, Codex, or Gemini CLI and syncs only usage numbers, keeping code and prompts private. It targets developers curious how their country compares on AI coding adoption.
TokenNations is a leaderboard ranking countries by aggregate AI coding tool token usage instead of individuals. A CLI reads local usage logs from tools like Claude Code, Codex, or Gemini CLI and syncs only usage numbers, keeping code and prompts private. It targets developers curious how their country compares on AI coding adoption.