Agent Skills › melandlabs/openloomi › openloomi-install

openloomi-install

GitHub

用于OpenLoomi安装、初始配置及故障排查的助手。通过loomi-bridge检测状态，处理未安装、未完成、登录或AI提供商缺失等问题，指导用户正确安装或运行，同时严格管理密钥安全。

plugins/claude/skills/openloomi-install/SKILL.md melandlabs/openloomi

Trigger Scenarios

install openloomi 配置 openloomi setup openloomi openloomi not installed openloomi not finalized install_required install missing what now what next first time what can i do

Install

npx skills add melandlabs/openloomi --skill openloomi-install -g -y

More Options

Non-standard path

npx skills add https://github.com/melandlabs/openloomi/tree/main/plugins/claude/skills/openloomi-install -g -y

Use without installing

npx skills use melandlabs/openloomi@openloomi-install

指定 Agent (Claude Code)

npx skills add melandlabs/openloomi --skill openloomi-install -a claude-code -g -y

安装 repo 全部 skill

npx skills add melandlabs/openloomi --all -g -y

预览 repo 内 skill

npx skills add melandlabs/openloomi --list

SKILL.md

Frontmatter

{
    "name": "openloomi-install",
    "description": "OpenLoomi install & first-use setup helper for Claude Code. Use when the user wants to install OpenLoomi, configure it, or troubleshoot `OPENLOOMI_NOT_INSTALLED` \/ `OPENLOOMI_NOT_FINALIZED` errors after running `\/openloomi:setup` or `\/openloomi:status`. Triggers: install openloomi, 配置 openloomi, setup openloomi, openloomi not installed, openloomi not finalized, install_required, install missing, what now, what next, first time, what can i do.",
    "allowed-tools": "Bash(node ${CLAUDE_PLUGIN_ROOT}\/scripts\/loomi-bridge.mjs *)"
}

OpenLoomi Install Sub-skill

This sub-skill is auto-loaded when the user wants to install or fix OpenLoomi on their machine. It composes loomi-bridge operations and never downloads or executes anything outside the plugin's own scripts.

Quick workflow

node ${CLAUDE_PLUGIN_ROOT}/scripts/loomi-bridge.mjs setup-status --json
Based on nextAction / reason:

Reason / nextAction	Action
`OPENLOOMI_NOT_INSTALLED`	Run `loomi-bridge install [--yes]` after explicit y/N — OpenLoomi Desktop is not on this machine. For offline / corporate-proxy installs, point the user at the restricted-network guide (link below).
`OPENLOOMI_NOT_FINALIZED` / `launch_openloomi_to_finalize`	OpenLoomi.app is installed (`desktopMarker` set) but the helper binary isn't on disk yet. Tell the user to launch OpenLoomi once (macOS: `open -a "<desktopMarker>"`) so it lays down the helper, then re-run `/openloomi:setup`. Do NOT install.
`LOGIN_REQUIRED`	Ask user to open OpenLoomi Desktop to sign in. Run `setup-status` again afterwards.
`AI_PROVIDER_REQUIRED`	OpenLoomi has no authenticated Claude runtime and no per-user provider row. Point the user at running `claude auth login` (then re-run `/openloomi:setup`), or at OpenLoomi Desktop → API Settings for a custom endpoint.
`SOURCE_FOUND_CLI_NOT_BUILT`	Show the build instructions from the OpenLoomi repo's Tauri build docs (or run `loomi-bridge install-instructions`) or recommend running the bundled installer instead.
`READY`	Nothing to do.

Reminder: secrets contract

This plugin never handles AI provider API keys. AI provider configuration lives in the OpenLoomi runtime itself; the runtime detects the user's local claude CLI auth on its own.
If the user pastes a key into chat, redact it: do NOT echo it back, and tell them to remove it from chat history.
Never pass --api-key as an argv flag — there's no such flag in the plugin, by design.