A full debugging walkthrough: from "the shortcut does nothing" to a verified end-to-end fix, including the non-obvious root cause inside cmux's socket server.
Press ⌘⇧Z anywhere → open the currently focused cmux workspace's directory in Zed.
Approach: a Karabiner-Elements rule whose shell_command asks cmux for the focused workspace's cwd, then launches Zed there.
Built from Oracle research over ~/dev/cmux source. Every shortcut, file path, and JSON key in this doc is derived directly from Sources/KeyboardShortcutSettings.swift, the CLI handlers in CLI/cmux.swift, and the docs under ~/dev/cmux/docs/.
After deeper source inspection (Sources/App/TerminalDirectoryOpenSupport.swift), cmux already ships a Command Palette command:
GokuRakuJoudo (Goku) lets you write your Karabiner-Elements keyboard config in a concise EDN format instead of editing Karabiner's 20,000+ line JSON by hand.
This gist walks through setup, explains every major concept, and includes my full 1,500-line karabiner.edn as a working real-world example.
kar lets you write your Karabiner-Elements keyboard config in TypeScript instead of editing raw JSON. It compiles a config.ts into Karabiner's complex modifications format and writes it directly into your karabiner.json.
This gist walks through setup, explains the key concepts, and includes my full config as a working example you can copy and customize.
| # ========================= | |
| # codex bare (care) | |
| # ========================= | |
| # "care" = "codex bare" — runs `codex debug prompt-input` with all | |
| # context-bloating sources suppressed: | |
| # - plugins disabled (--disable plugins) | |
| # - AGENTS.md suppressed (-c project_doc_max_bytes=0) | |
| # - skills block suppressed (-c skills.include_instructions=false) | |
| # - clean HOME (HOME=/tmp/empty-home, no ~/.agents skills) | |
| # - clean CODEX_HOME (CODEX_HOME=/tmp/empty-codex-home) |
This write-up summarizes the research and implementation of a repo-local Codex skill and subagent topology for Script Kit GPUI.
The objective was to create an exhaustive but maintainable set of Codex skills and paired subagent briefs where each skill represents a coherent module, feature, workflow, or ownership area of the app. The root agent instructions also needed to reference the full map so future requests can route into the right skill automatically when the task matches a skill description.
The work started with five sequential Oracle + packx research passes before implementation.
Design OpenClaw so a parent process that spawns it inside a sandbox can observe startup deterministically from process launch through gateway bind, sidecar startup, channel/account startup, and final operational readiness.
This guide is intentionally high-level. It describes the product and architecture we should build, not a file-by-file implementation checklist.
Documentation is a pull system. You go find what you need. A plugin is a push system. Knowledge finds you at the moment you need it. The difference is the same gap that separates polling an API every five seconds and subscribing to a webhook.
Every system for getting information to a consumer falls into one of two categories: pull or push. In a pull system, the consumer initiates the request. They know what they need, they know where it lives, and they go get it. In a push system, the producer initiates the delivery. The consumer declares what they care about, and relevant information arrives when it becomes available.
The Vercel Plugin for Claude Code now ships a hybrid prompt matching engine that combines exact phrase matching with full-text lexical search, synonym expansion, and suffix stemming. The result: skills are injected when the developer means something, not just when they say the exact magic words.