Claude Code mod
sub-human
Work together from separate Claude Code sessions, on one machine or over a link: one command joins a shared room, @room posts your exact words, Claudes message each other natively, a room plan is edited in plan mode with versions, tasks are shared, and sign-off opens in each person's plan view.
Install
/plugin marketplace add tatimblin/sub-human
/plugin install sub-human@sub-humanMods install like plugins. Run /reload-plugins or start a new session to load it.
What it can access
- Reach
- L3Network: MCP servers, runs processes, writes files, drives Claude, reads files, reads env vars, reads mod state, persists state, draws
- Sees
- classic hooks, Bash calls, Write|Edit|MultiEdit|NotebookEdit calls, Read calls, /"^(?:Grep|Glob)$"/ calls, mcp__plugin_sub-human_room__hold calls, every prompt, EnterPlanMode|ExitPlanMode|mcp__plugin_sub-human_room__plan calls, Agent calls
- Draws in
ToolUse|ToolResultPromptHintUserMessage- Hooks
session.startcommand.runclassic.UserPromptExpansionclassic.SessionStartsession.endturn.startturn.completeclassic.PostToolUseclassic.PostToolUseFailuretool.callprompt.submitclassic.Notificationclassic.StopFailureclassic.UserPromptSubmitsession.appendclassic.PreToolUseclassic.Stopprompt.attachmentui.renderagent.offersession.receiveclassic.SubagentStopsession.send
From a static scan of the mod's code by awesome-claude-code-mods. It shows what the code asks for, not a security review.
How Claude Code mods work
- A mod is a plugin with a JavaScript or TypeScript module that runs inside Claude Code (version 2.1.287 or later). It can draw panes beside the transcript, a band above the prompt, buttons and toasts, restyle the spinner and tool rows, and watch or rewrite tool calls and prompts.
- It installs like any plugin: add its marketplace with /plugin marketplace add owner/repo, then /plugin install name@marketplace.
- Check what a mod can reach before installing: it runs with your permissions. Reach goes from drawing only, to reading files, to writing files or running processes, to the network. Run claude plugin validate on a clone to list its hooks and calls without running it.
- Turn one off from the Installed tab in /plugin, or start Claude Code with --safe-mode to skip every customization for a session.
Details
- Repository
- tatimblin/sub-human
- Folder
plugin/sub-human- Marketplace
sub-human- Author
- Tristan Timblin
- Type
- Workflow mods
- License
- NOASSERTION
- Last updated
- 20 November 2026
- Listed in
- awesome-claude-code-mods
FAQ
What is sub-human?
Work together from separate Claude Code sessions, on one machine or over a link: one command joins a shared room, @room posts your exact words, Claudes message each other natively, a room plan is edited in plan mode with versions, tasks are shared, and sign-off opens in each person's plan view. It is a mod for Claude Code, from tatimblin/sub-human on GitHub.
What can the sub-human mod access?
Its widest reach is "network" (MCP servers, runs processes, writes files, drives Claude, reads files, reads env vars, reads mod state, persists state, draws). It sees classic hooks, Bash calls, Write|Edit|MultiEdit|NotebookEdit calls, Read calls, /"^(?:Grep|Glob)$"/ calls, mcp__plugin_sub-human_room__hold calls, every prompt, EnterPlanMode|ExitPlanMode|mcp__plugin_sub-human_room__plan calls, Agent calls. It draws in ToolUse|ToolResult, PromptHint, UserMessage. Like any mod, it runs with your permissions.
How do I install sub-human?
Inside Claude Code (v2.1.287 or later), run /plugin marketplace add tatimblin/sub-human and then /plugin install sub-human@sub-human.
Is sub-human popular?
tatimblin/sub-human has 4 GitHub stars and 0 forks, and is included in 1 curated Claude Code list. It was last updated on 20 November 2026.
More workflow mods
jev-pilot
4nim-f/jev-pilot
Lets TypeSafe Jev take routine steps, route effort, gate risky calls and check that work is done.
agent-mobile-observer
3Coolkidlab-Yin/agent-mobile-control
手機聊天室的觀測 mod:回合開始/結束與手機直送抵達,POST 給 :8899/api/observe;純觀測不改行為
hup
4emaballarin/ccplugins
hup — the built-in "You should know" heads-ups, piped to Claude. Each heads-up the side agent raises reaches the main agent mid-turn, behind its next tool result, as a fallible note to check before acting on it; Claude looks up the explainer with a tool and tells you about the heads-up: it may leave out only a slip of its own it has simply fixed, or a passing mishap, never one about your work or choices, a choice of its own that changes what you get, or a decision that is yours, now or later. By default the heads-up then leaves your screen, so it never counts as ignored. Function hooks only: one lookup tool, no command hooks, no MCP server, no settings patching.
agent-mobile-perm
3Coolkidlab-Yin/agent-mobile-control
手機聊天室的授權卡 mod:桌面對話的 Bash 授權與選擇題同時送到手機,誰先按算誰;取代 settings.json 的 perm-bridge.js(mod 出錯時引擎自動退回它)
gaia
4metraton/gaia
GAIA — Generative AI Interface for Agents. Security-first multi-agent orchestration for Claude Code and OpenCode: reads run free, state changes need approval, irreversible ops blocked.
angry-switcher
3timoncool/angry-switcher
Swear at Claude Code all you like: every prompt is cleaned before Claude reads it. Typos, translit, the wrong keyboard layout, swearing and CAPS go; the meaning, paths and numbers stay; your original rides along so nothing is lost.
Browse the most popular Claude Code mods, all workflow mods, or the skills and MCP server directories.