菜单
Control Supacode from the terminal. Use when running Supacode CLI commands, managing worktrees, tabs, and surfaces programmatically, or when inside a Supacode terminal session.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/Sheshiyer/skill-clusters --skill supacode-cli命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | supacode-cli |
| description | Control Supacode from the terminal. Use when running Supacode CLI commands, managing worktrees, tabs, and surfaces programmatically, or when inside a Supacode terminal session. |
| version | 1.0.0 |
| cluster | agentic-ops |
Control Supacode from the terminal. The supacode command is available in all Supacode terminal sessions.
NEVER call supacode tab new or supacode surface split without capturing
the output. They print the new UUID to stdout. Without it you cannot target
the resource afterward.
NEVER omit -t/-s when targeting a created resource. The env vars point
to your own shell, not to anything you created.
For new tabs, surface ID = tab ID.
TAB_ID=$(supacode tab new -i "npm start")
SPLIT_ID=$(supacode surface split -t "$TAB_ID" -s "$TAB_ID" -d v -i "npm test")
supacode surface close -t "$TAB_ID" -s "$SPLIT_ID"
supacode tab close -t "$TAB_ID"
supacode tab new -i "npm start" # BAD: not captured
supacode surface split -d v -i "test" # BAD: missing -t/-s, targets your shell
supacode worktree [list [-f]|focus|run [-c]|stop [-c]|script list|archive|unarchive|delete|pin|unpin] [-w <id>]supacode tab [list [-w] [-f]|focus|new|close] [-w <id>] [-t <id>] [-i <cmd>] [-n <uuid>]supacode surface [list [-w] [-t] [-f]|focus|split|close] [-w <id>] [-t <id>] [-s <id>] [-i <cmd>] [-d h|v] [-n <uuid>]supacode repo [list | open <path> | worktree-new [-r <id>] [--branch] [--base] [--fetch] [--name] [--location]]supacode settings [<section>]supacode socketlist outputs one ID per line (percent-encoded for worktrees/repos, UUIDs for tabs/surfaces).
worktree script list outputs tab-separated <uuid>\t<kind>\t<displayName> rows; running scripts are ANSI-underlined.
Use these IDs directly as -w, -t, -s, -r, -c flag values.
Flags: -w (worktree), -t (tab), -s (surface), -r (repo), -c (script UUID for worktree run/stop), -i (input), -d (direction), -n (new ID).
Env var defaults only target your own shell session. Pass explicit IDs for created resources.
Arcplume runs Grok through the Grok Build CLI's own OAuth-authenticated session (grok login) for image generation, with strict preflight validation, secret-safe handling, and headless CLI-driven execution -- no separate XAI_API_KEY billing. Video falls back to the billed xAI API. USE WHEN a user wants to generate an image via a locally logged-in Grok Build CLI session, e.g. 'generate an image with grok', 'use grok build', or 'use my logged-in grok session'.
Shared reference for the Selemene cluster: the two report surfaces (deterministic Rust reports vs. narrative witness-pipeline readings), the @selemene/bridge CLI contract, the output manifest format, and non-prescriptive witnessing tone. USE WHEN deciding which Selemene surface to invoke or when routing between birth/compatibility/transit reports and solo/dyadic readings.
Route Selemene Engine tasks to the right surface: deterministic reports (birth/compatibility/transit) via selemene-core and the @selemene/bridge CLI, or narrative witness readings via selemene-report. USE WHEN the user wants anything under the Selemene/Noesis umbrella but has not named the exact surface.