electron-pro · qdhenry — Claude Code subagent: what it does, how to install
A Claude subagent that runs an Electron project written in TypeScript: main and renderer processes, IPC through a preload bridge, packaging with electron-builder.
no license ★ 1,357 updated 2026-03-01
.claude/agents/electron-pro.mdWhat electron-pro does
It splits an Electron app into the parts that matter: a main process, a renderer, and a preload script that hands over a narrow API through contextBridge. Channels carry TypeScript types, so a mismatch surfaces at build time. Security gets the same treatment: context isolation on, sandboxing, a restrictive CSP, and no Node.js inside windows that load remote pages. Native work such as file access and dialogs stays in the main process.
The tool list is broad: file reads and edits, Bash, glob and grep, web search and fetch. Two MCP servers back the thinking side — context7 for library docs, sequential-thinking for architecture calls. It runs on sonnet. Work goes in a plain order: scaffold the folders and a strict tsconfig.json first, then typed IPC, tests, electron-builder targets and code signing.
Behaviour checks go to Playwright end-to-end tests rather than hand clicking. Its tool list names mcp__context7__get-library-docs and mcp__sequential-thinking__sequentialthinking; skip connecting those MCP servers and the library lookups plus the separate reasoning pass are not available to it.
What it helps with
- scaffold an Electron + TypeScript project from scratch
- type the IPC channels between main and renderer
- lock down context isolation and CSP before release
- package builds for several platforms with
electron-builder - cover a window with a Playwright end-to-end test
Example prompts for the agent
- Scaffold an Electron + TypeScript app with separate main, renderer and preload folders and a strict
tsconfig.json. - Wire IPC between renderer and main through
contextBridgeand type the channels. - Set up
electron-builderfor Windows and macOS and add a Playwright end-to-end test.
How to install electron-pro
Manually: put the files in place
.claude/agents/electron-pro.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/electron-pro.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/electron-pro.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about electron-pro
How do I install electron-pro in Claude Code?
Copy electron-pro.md from the qdhenry/Claude-Command-Suite repo into .claude/agents/electron-pro.md in your project.
What does electron-pro do?
It works on Electron desktop apps written in TypeScript: process layout, IPC, window security, tests and packaging.
Does it need the context7 and sequential-thinking MCP servers?
Both appear in its tool list. With them connected it can pull library docs and use a separate reasoning pass; without them those calls are unavailable.
Similar subagents
All subagents for Agent setup: memory, context, orchestration →electron-pro by other authors comes first.