Agent skill

Night Autopilot

by cat-xierluo in cat-xierluo/legal-skills

无人值守自主推进模式(宿主无关、领域无关)——定时触发下,Agent 自动整理任务、查找下一项、完成或派发、监测 subagent/独立 worker、验收推进并记录交接。适用于代码项目(PR/测试验收)与法律项目(文书核对+律师终审门)等任何有任务源和可定义验收判据的领域。在用户说「我要去睡觉了/今晚继续推进/不间断运行/明早给我结果」并要求定时派发验收时使用。不要用于:单次后台任务(直接派发…

MITAuto-check passedAgent Workflows

Install Night Autopilot

skills CLI
$ npx skills add cat-xierluo/legal-skills --skill night-autopilot -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install cat-xierluo/legal-skills night-autopilot --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/cat-xierluo/legal-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/night-autopilot .claude/skills/night-autopilot && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
night-autopilot
GitHub stars
720
Token cost
~963 tokens
SKILL.md length
222 words
Files
2
Skills in repo
62
Repo updated
First seen
Licence
MIT

At a glance

无人值守自主推进模式(宿主无关、领域无关)——定时触发下,Agent 自动整理任务、查找下一项、完成或派发、监测 subagent/独立 worker、验收推进并记录交接。适用于代码项目(PR/测试验收)与法律项目(文书核对+律师终审门)等任何有任务源和可定义验收判据的领域。在用户说「我要去睡觉了/今晚继续推进/不间断运行/明早给我结果」并要求定时派发验收时使用。不要用于:单次后台任务(直接派发…

  • Works in 8 steps: 核心循环(五步,宿主无关) → 定时触发层(与循环解耦) → 启动前核对(任一不满足不进入) → …
  • Tasks that involve Multi-agent orchestration
  • SKILL.md covers 1. 核心循环(五步,宿主无关), 2. 定时触发层(与循环解耦), 3. 启动前核对(任一不满足不进入) and 4. 验收判据(领域参数化), plus 4 more sections
  • Calls claude, codex and gh

What it does

Night Autopilot is an agent skill from cat-xierluo/legal-skills. 无人值守自主推进模式(宿主无关、领域无关)——定时触发下,Agent 自动整理任务、查找下一项、完成或派发、监测 subagent/独立 worker、验收推进并记录交接。适用于代码项目(PR/测试验收)与法律项目(文书核对+律师终审门)等任何有任务源和可定义验收判据的领域。在用户说「我要去睡觉了/今晚继续推进/不间断运行/明早给我结果」并要求定时派发验收时使用。不要用于:单次后台任务(直接派发即可)、用户在场的普通编排(用 multi-agent-orchestration)、一次性定时提醒。

Its SKILL.md is about 960 tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `CHANGELOG.md`).

It sits in Agent Workflows, covering Multi-agent orchestration and Subagents. The licence is MIT.

When your agent uses it

  • Tasks that involve Multi-agent orchestration
  • Tasks that involve Subagents

Example prompts

  • “/night-autopilot”

Workflow steps

8 steps, taken from the step headings in SKILL.md.

  1. 核心循环(五步,宿主无关)
  2. 定时触发层(与循环解耦)
  3. 启动前核对(任一不满足不进入)
  4. 验收判据(领域参数化)
  5. 巡检与 worker 监测
  6. 队列锚点模板(标准化)
  7. 晨间交接
  8. 边界与已知限制

What it can do on your machine

Read from SKILL.md and the folder at commit db2c58c. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • claude
    • codex
    • gh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Night Autopilot loads about 963 tokens when it runs. Until then it costs about 67 tokens; SKILL.md has 222 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~67
When it runs · the whole SKILL.md, loaded when a task matches
~963

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from cat-xierluo/legal-skills at commit db2c58c, republished under its MIT licence (© cat-xierluo). 222 words, ~963 tokens.

Download SKILL.mdSave it as .claude/skills/night-autopilot/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
night-autopilot
description
无人值守自主推进模式(宿主无关、领域无关)——定时触发下,Agent 自动整理任务、查找下一项、完成或派发、监测 subagent/独立 worker、验收推进并记录交接。适用于代码项目(PR/测试验收)与法律项目(文书核对+律师终审门)等任何有任务源和可定义验收判据的领域。在用户说「我要去睡觉了/今晚继续推进/不间断运行/明早给我结果」并要求定时派发验收时使用。不要用于:单次后台任务(直接派发即可)、用户在场的普通编排(用 multi-agent-orchestration)、一次性定时提醒。
homepage
https://github.com/cat-xierluo/legal-skills
author
杨卫薪律师(微信ywxlaw)
version
0.2.1
license
MIT

Night Autopilot(无人值守自主推进)

把「用户离开、Agent 通宵推进」固化为可复用循环。v0.2.0 起宿主无关、领域无关:定时只是触发器,循环本体是五步通用流程;验收判据按领域参数化(代码型 / 法律型)。

首战实例:dsh-plugins 仓 2026-09-30 夜(ZCode 宿主、15 分钟轮、代码域、六卡交付);泛化需求来自用户同日指示(其他宿主定时机制不同 + 法律项目也要能夜跑)。

定位与分工:本 skill 是自主推进循环层——任务自组织、定时触发、巡检验收、记录交接。通道选择、派发门禁、worker 纪律以其宿主规则 multi-agent-orchestration 为准;Git/PR 规则用 git-workflow。本 skill 不重复它们的内容。

1. 核心循环(五步,宿主无关)

  1. 整理任务:从任务源自动聚合生成/刷新候选队列——项目 TASKS 文件、issue、会话遗留待办、用户口述。去重、按依赖排序;价值判据自动分桶:可派(实现/验证类,交 worker)、Agent 自留(研究/调查类)、需用户(决策/在场/授权类,只登记不推进);存疑项标记待确认,不猜。
  2. 查找下一项:按桶取件——可派且活跃 worker 未满(默认 ≤3)则派发;只有自留项则本循环自己做;只剩需用户项则转纯巡检记录。
  3. 完成或派发:派发按 orchestration 通道判据选型(中小卡宿主 subagent,独立 worker 场景走其模式表);任务卡自包含(目标/allowed/禁止项/验收命令/交付协议)。
  4. 验收推进:按领域验收判据(§4)核验交付,过则收口(合并/归档/登记),不过则小修闭环或泊车——无人值守期宁保守。
  5. 记录与交接:每轮更新队列锚点(状态机:完成/在途/待派/泊车);结束时出三类交接报告(完成/泊车/待用户)。

2. 定时触发层(与循环解耦)

循环本体不含任何定时机制;触发器契约:输入 = 队列锚点路径 + 安全边界(谁可写/禁碰清单)+ 循环 prompt;输出 = 锚点状态写回 + 交接产物。满足契约的任意调度器都可驱动。

宿主定时驱动状态
ZCode内建定时任务(CronCreate/CronUpdate,prompt 注入本循环)首战实测
Claude Code原生 /goal 模式(目标驱动自主迭代,progress/audit 双面)优先;次选 headless CLI(claude -p --continue)+ 系统调度待实测(映射规范见本地 TASKS NA-D6)
Codexgoal mode(迭代下一工作单元直至目标达成)优先;次选无头 CLI(codex exec)+ 系统调度待实测
OS 级兜底cron / launchd 调宿主 CLI 无头入口,参数喂循环 prompt 与锚点路径任何宿主可用

兜底路径要点:调度器只负责「到点拉起 Agent + 传参」,循环状态全部落锚点文件(进程无常驻假设);每轮超时自限(如 15 分钟 kill),防调度堆积。

3. 启动前核对(任一不满足不进入)

#前提说明
1用户明确授权无人值守 + 频率 + 大致时长;无授权不得自设定时任务
2任务源可读队列可自动聚合(§1.1),但安全边界必须成文:可写范围、禁碰清单(生产环境/他人领地/敏感数据)、交付通道
3验收判据成文按领域选型(§4)并在锚点声明;法律域必须显式列出人工终审门
4恢复锚点落盘队列状态 + rails + 停止条件写入持久文件(会话上下文可能被摘要/重启)
5触发器就绪按宿主表建定时任务;睡前报告(已派什么/护栏/醒来会看到什么)

4. 验收判据(领域参数化)

代码型(默认):全套测试绿 + lint/机械审查 0 FAIL + diff 在 allowed 范围 + NOT_VERIFIED 如实标注。交付形态 = 分支 + PR(Git 规则走 git-workflow)。

法律型(文书/检索类任务):

  • 机械可验层(Agent 可自动核):核对清单逐项(文书要件齐全)、引用核验(法条/案号与权威库比对——工具如元典幻觉校验)、当事人与案号信息一致性、格式与期限计算复核(日期/金额/次数)、版本与留痕(文件名/修改记录)。
  • 人工终审门(不可自动越过):律师终审、签署、对外发送(法院/对方/客户)一律留给用户——夜间产出只能是「初稿 + 核对记录 + 晨间审阅清单」。
  • 交付形态 = 文件 + 核对记录(不强制 git/PR;有仓则走仓)。
  • 数据红线:实际客户材料默认不进无人值守循环——脱敏/合成先行,涉密材料永远在场处理。

自定义领域按同构扩展:定义「机械可验层 + 人工终审门 + 交付形态」三元组即可。

5. 巡检与 worker 监测

  • 双通道:宿主 subagent 与独立 worker(Orca supervised/terminal-managed/tmux)——完成权威与巡检证据优先级按 orchestration §2 模式表执行,本 skill 不另造。
  • 死 worker 判定(双证,2026-09-30 两次实战提炼):①会话/输出记录停更(转录 mtime 长期不变)②worktree/产物无进展(文件 mtime 停滞、无 commit)。双证齐 → 判死;单证不判(可能在跑长测试)。
  • 处置分级:完整交付未推送 → PM 独立复验后代交付;半途 WIP → 清理重派(任务卡可重放);同卡连续 2 次失败 → 泊车记录,不盲目重拉。
  • 僵尸 UI 条目:界面显示运行中但任务系统查无此任务——用停止接口验证(返回 No task found 即纯显示残留,刷新即消)。
  • 基础设施故障回退:如 GitHub GraphQL 建 PR 持续报错 → REST API(gh api repos/<o>/<r>/pulls)。

6. 队列锚点模板(标准化)

markdown
## 授权与 rails
授权(日期+频率+时长)/ 定时任务 id / 领域与验收判据型 / 禁碰清单 / 交付形态 / 停止条件

## 队列状态(每轮更新)
1. [完成·PR/交付物] 卡名(去向)
2. [在途·workerX] 卡名(要点 + allowed)
3. [自留] 研究项
4. [待派] 卡名(派发前置)
5. [泊车] 卡名(原因 + 证据)
6. [需用户] 事项

## 巡检日志
- HH:MM 轮:一行结论(做了什么/全静/泊车原因)

7. 晨间交接

  • 三类汇总:已完成(交付物清单+去向)、泊车(原因+证据)、待用户决策。
  • 提醒用户停止定时任务(或用户说停即删);清理一次性资源。

8. 边界与已知限制

  • 触发器宿主表中 Claude Code(/goal)/ Codex(goal mode)两行为待实测——实测后更新本表与映射规范(NA-D6)。原生模式只替代循环驱动,不替代锚点、验收判据与人工终审门。
  • 法律域机械可验层是框架不是清单——具体核对项按项目定制,且永不把人工终审门自动化。
  • 会话绑定的内建定时器随会话终止而停;跨会话持久调度用 OS 级兜底。
  • 额度共享(宿主 subagent 与主会话);每轮时长自限,不做无界长任务。

© cat-xierluo, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/night-autopilot of cat-xierluo/legal-skills.

  • SKILL.md
  • CHANGELOG.md

Open the folder on GitHubat commit db2c58c

Compare with similar skills

Night Autopilot next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Night Autopilot compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Night Autopilot this skillcat-xierluo/legal-skills720—~963Automated safety check: PassMIT
Paseo Advisor Second Opiniongetpaseo/paseo20k1 repos~756Automated safety check: PassCustom licence
O2 Review Loopopenobserve/openobserve22k—~3.7kAutomated safety check: PassAGPL-3.0
Harness Agent Team Designerrevfactory/harness9.1k—~4.5kAutomated safety check: PassApache-2.0
ClawTeam Multi-Agent Swarmwin4r/ClawTeam-OpenClaw1.5k1 repos~2.9kAutomated safety check: PassMIT
Sub-Agent Delegationcodewhale-hq/Codewhale41k—~790Automated safety check: PassMIT

Similar skills

  • Launches one separate agent through Paseo to give a second opinion on the current task, with a self-contained briefing and no permission to edit files.

    20k GitHub starsUsed in 1 repo~756 tokens
    Agent WorkflowsAuto-check passed
  • O2 Review Loop

    openobserve/openobserve

    Splits a change into planner, coder and independent reviewer roles: you confirm a spec, a subagent implements it, and a separate reviewer checks each round's local WIP commit.

    22k GitHub stars~3.7k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Harness Agent Team Designer

    revfactory/harness

    Designs a project-specific agent harness: defines specialist agents, writes the skills they follow, picks an execution mode and model for each, and keeps the setup maintained.

    9.1k GitHub stars~4.5k tokensUpdated 13 days ago
    Agent WorkflowsAuto-check passed
  • ClawTeam Multi-Agent Swarm

    win4r/ClawTeam-OpenClaw

    Launches a swarm of specialist Hermes agents in git-worktree-isolated tmux windows with a kanban board and file-based inboxes, using built-in templates like hedge-fund and code-review.

    1.5k GitHub starsUsed in 1 repo~2.9k tokens
    Agent WorkflowsAuto-check passed
  • Sub-Agent Delegation

    codewhale-hq/Codewhale

    Guides when and how to split multi-step coding, research or verification work into focused sub-agent runs while the parent keeps integration and final checks.

    41k GitHub stars~790 tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Sets up and drives Ruflo, an npm-installed orchestration layer for multi-agent swarms, persistent memory, routing, hooks and its MCP tool catalog.

    74k GitHub starsUsed in 1 repo~975 tokens
    Agent WorkflowsAuto-check passed

More from cat-xierluo/legal-skills

All 62 skills in this repo
  • Elements-Style Complaint Generator

    cat-xierluo/legal-skills

    Converts a lawyer's ordinary complaint or a described case into the Supreme People's Court's elements-style Word template, with layout checks on the result.

    721 GitHub stars~2.6k tokensUpdated yesterday
    Auto-check: notes
  • Lecture Performance Review

    cat-xierluo/legal-skills

    Analyzes raw lecture transcripts for verbal tics, pacing, time use and promise follow-through, with optional slide-by-slide comparison and cross-session tracking.

    721 GitHub stars~2.2k tokensUpdated yesterday
    Auto-check passed
  • De-AI Polish for Chinese Articles

    cat-xierluo/legal-skills

    Detects and rewrites machine-sounding patterns in the body text of Chinese articles while keeping the author's facts, headings and legal terms intact.

    721 GitHub stars~2.9k tokensUpdated yesterday
    Auto-check passed
  • GitHub Star Manager

    cat-xierluo/legal-skills

    Finds GitHub projects mentioned in articles or screenshots and stars them, tracks updates to your starred repos, and builds an HTML dashboard to browse them.

    721 GitHub stars~2k tokensUpdated yesterday
    Auto-check: notes
  • Legal Harness Initializer

    cat-xierluo/legal-skills

    Sets up or incrementally updates AGENTS.md and CLAUDE.md for legal professionals, with a minimal safety baseline and a check that a new session loads and follows the rules.

    721 GitHub stars~2.3k tokensUpdated yesterday
    Auto-check passed
  • Moot Court Simulation Builder

    cat-xierluo/legal-skills

    Chinese-language skill that organizes a case file into a multi-role mock trial with judge, parties and clerk, producing a transcript, issue review and a to-strengthen list.

    721 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Night Autopilot

What does Night Autopilot do?

无人值守自主推进模式(宿主无关、领域无关)——定时触发下,Agent 自动整理任务、查找下一项、完成或派发、监测 subagent/独立 worker、验收推进并记录交接。适用于代码项目(PR/测试验收)与法律项目(文书核对+律师终审门)等任何有任务源和可定义验收判据的领域。在用户说「我要去睡觉了/今晚继续推进/不间断运行/明早给我结果」并要求定时派发验收时使用。不要用于:单次后台任务(直接派发…. Night Autopilot is an agent skill from cat-xierluo/legal-skills.

When should I use Night Autopilot?

Night Autopilot fits situations like: tasks that involve Multi-agent orchestration; tasks that involve Subagents.

How do I install Night Autopilot in Claude Code?

Run `npx skills add cat-xierluo/legal-skills --skill night-autopilot -a claude-code`. Or copy the skill folder (skills/night-autopilot in cat-xierluo/legal-skills) into .claude/skills/night-autopilot in your project. Claude Code loads it when a task matches its description.

How do I install Night Autopilot in Codex?

Run `npx skills add cat-xierluo/legal-skills --skill night-autopilot -a codex`. Or copy the skill folder (skills/night-autopilot in cat-xierluo/legal-skills) into .agents/skills/night-autopilot in your project. Codex loads it when a task matches its description.

Can I use Night Autopilot in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add cat-xierluo/legal-skills --skill night-autopilot -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/night-autopilot, .gemini/skills/night-autopilot, .github/skills/night-autopilot and .opencode/skills/night-autopilot in your project.

What does Night Autopilot need to run?

Going by SKILL.md and its folder, Night Autopilot needs the command-line tools its instructions call (claude, codex and gh).

Does Night Autopilot access the network?

SKILL.md contains no URLs. Its commands use gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Night Autopilot safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Night Autopilot use?

Night Autopilot is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Night Autopilot use?

About 963 tokens (SKILL.md is roughly 3.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Night Autopilot?

Skills that share tags, products or a category with Night Autopilot: Paseo Advisor Second Opinion (getpaseo/paseo, 20k stars), O2 Review Loop (openobserve/openobserve, 22k stars), Harness Agent Team Designer (revfactory/harness, 9.1k stars) and ClawTeam Multi-Agent Swarm (win4r/ClawTeam-OpenClaw, 1.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Night Autopilot?

cat-xierluo (a GitHub user) maintains it in cat-xierluo/legal-skills, which has 720 GitHub stars. The repository holds 62 skills in this directory. The repository was last updated on October 10, 2026.

Source: cat-xierluo/legal-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.