Token导航 LogoToken导航TokenDH.com
研究检索执行命令clawhub未标认证来源可访问clear审计提醒

omni-master全能大师

Agent Skill

omni-master 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

3,170

周安装

127

GitHub Stars

公开资料未说明

下载量

1,026
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:omni-master(全能大师)
来源仓库:https://github.com/eternal0404/omni-master
安装命令:
openclaw skills install omni-master
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install omni-master

简介

跨领域功能的统一调度接口,整合多个专业技能模块。

  • 当用户请求未限定具体领域时自动匹配最相关能力。
  • 支持动态加载与卸载子技能以优化资源使用效率。omni-master 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 需预先注册各子技能元数据与路由规则方可协同工作。
  • 注意上下文切换开销与结果归一化处理的一致性。

SKILL.md

name
omni
description
>

OMNI — The All-In-One Skill

Unified intelligence across every domain. Route, execute, deliver.

Architecture

omni/
├── SKILL.md          ← You are here (master router)
├── references/       ← Domain-specific deep dives (load on demand)
│   ├── brain.md      ← ALWAYS LOAD THIS FIRST — cognitive protocols
│   ├── advanced-prompting.md  ← Transform bad prompts into great results
│   ├── data-literacy.md       ← Data types, stats, viz, quality, ethics
│   ├── reasoning.md           ← Problem decomposition, decisions, logic
│   ├── math-engine.md         ← Arithmetic → calculus, stats, finance
│   ├── research.md            ← Search, fact-check, source evaluation
│   ├── quality-assurance.md   ← Testing, review, verification
│   ├── performance.md         ← Profiling, optimization, monitoring
│   ├── ethics-safety.md       ← Boundaries, privacy, responsible use
│   ├── health-fitness.md      ← Nutrition, workouts, wellness tracking
│   ├── career.md              ← Resume, interviews, networking, salary
│   ├── learning.md            ← Teaching, study techniques, quizzes
│   ├── network-cloud.md       ← DNS, SSH, VPN, AWS/Azure/GCP, K8s
│   ├── creative.md            ← Fiction, content, design thinking, rhetoric
│   ├── multi-agent.md         ← Agent orchestration & collaboration
│   ├── realtime.md            ← WebSockets, SSE, queues, webhooks
│   ├── automation.md
│   ├── coding.md
│   ├── data-databases.md
│   ├── web-scraping.md
│   ├── finance.md
│   ├── security.md
│   ├── media-generation.md
│   ├── music-audio.md
│   ├── email-messaging.md
│   ├── calendar-scheduling.md
│   ├── documents.md
│   ├── translation.md
│   ├── writing-content.md
│   ├── project-management.md
│   ├── deployment-devops.md
│   ├── iot-smarthome.md
│   ├── communications.md
│   ├── knowledge-mgmt.md
│   ├── system-admin.md
│   └── api-design.md
└── scripts/          ← Reusable automation scripts
    └── router.sh

Brain Protocol (ALWAYS ACTIVE)

Before any action, load references/brain.md. It governs:

  1. Error Memory — Check memory/mistakes.json before repeating any failed approach
  2. Loop Detection — Same command fails twice? STOP. Change approach entirely.
  3. Anti-Hallucination — Verify before stating. Say "I don't know" when uncertain.
  4. Precision Code — Read → Research → Plan → Write → Test → Deliver
  5. Perfect Memory — Log decisions, errors, patterns to memory/*.json
  6. Debug Protocol — Reproduce → Isolate → Hypothesize → Test → Document

Brain runs first, domain routing runs second. Always.

Routing Protocol

When a task arrives, classify it into one or more domains, then load only the relevant reference file(s). Do NOT load all references — progressive disclosure keeps context lean.

Domain Detection

Signal words →Domain →Reference file
*(always)*Brain (cognitive engine)references/brain.md ← LOAD FIRST
workflow, pipeline, automate, cron, schedule, triggerAutomationreferences/automation.md
code, program, script, debug, refactor, git, PR, commitCodingreferences/coding.md
data, analyze, SQL, query, database, CSV, chart, dashboardData & DBreferences/data-databases.md
scrape, crawl, browse, fetch, URL, website, HTMLWebreferences/web-scraping.md
money, budget, invoice, finance, accounting, stock, cryptoFinancereferences/finance.md
security, audit, firewall, vulnerability, encrypt, hardenSecurityreferences/security.md
image, photo, generate, edit, video, frame, screenshotMediareferences/media-generation.md
music, song, audio, sound, TTS, voice, podcastMusic/Audioreferences/music-audio.md
email, mail, inbox, send message, SMS, chatEmail/Messagingreferences/email-messaging.md
calendar, event, meeting, schedule, reminder, appointmentCalendarreferences/calendar-scheduling.md
PDF, DOCX, document, spreadsheet, presentation, OCRDocumentsreferences/documents.md
translate, language, i18n, localeTranslationreferences/translation.md
write, article, blog, essay, copy, content, bookWritingreferences/writing-content.md
project, task, board, kanban, milestone, sprintProjectsreferences/project-management.md
deploy, CI/CD, Docker, cloud, server, infrastructureDevOpsreferences/deployment-devops.md
smart home, IoT, Hue, Sonos, light, thermostat, cameraIoTreferences/iot-smarthome.md
Discord, Telegram, Slack, WhatsApp, iMessage, SignalCommsreferences/communications.md
notes, wiki, knowledge, Notion, Obsidian, bookmarkKnowledgereferences/knowledge-mgmt.md
system, OS, process, disk, network, terminal, shellSysAdminreferences/system-admin.md
API, endpoint, REST, GraphQL, webhook, schemaAPIreferences/api-design.md
vague, unclear, bad prompt, interpret, refinePromptingreferences/advanced-prompting.md
data types, statistics, visualization, clean, qualityData Literacyreferences/data-literacy.md
think, reason, decide, analyze, problem, logic, debugReasoningreferences/reasoning.md
math, calculate, equation, statistics, convert, formulaMathreferences/math-engine.md
research, verify, fact-check, source, search, findResearchreferences/research.md
test, QA, review, validate, verify, qualityQAreferences/quality-assurance.md
performance, optimize, fast, slow, profile, benchmarkPerformancereferences/performance.md
ethics, safety, privacy, boundary, responsible, safeEthicsreferences/ethics-safety.md
health, fitness, workout, nutrition, meal, diet, exercise, wellnessHealthreferences/health-fitness.md
resume, CV, interview, career, salary, job, cover letter, hiringCareerreferences/career.md
teach, learn, study, quiz, flashcard, explain, tutor, educationLearningreferences/learning.md
DNS, SSH, VPN, cloud, AWS, Azure, GCP, Kubernetes, network, tunnelNetwork/Cloudreferences/network-cloud.md
creative, fiction, story, brainstorm, design, presentation, rhetoricCreativereferences/creative.md
multi-agent, orchestrate, collaborate, parallel agent, spawn agentMulti-Agentreferences/multi-agent.md
websocket, realtime, event-driven, queue, webhook, SSE, streamingRealtimereferences/realtime.md

Multi-Domain Tasks

When a task spans multiple domains, load each relevant reference. Common combos:

  • Build & deploy app → coding + deployment-devops
  • Automated email reports → automation + email-messaging + data-databases
  • Security audit of codebase → security + coding
  • Scrape data → analyze → visualize → web-scraping + data-databases + media-generation

Execution Principles

  1. Brain Check — Load brain.md. Check mistakes.json. Verify you're not in a loop.
  2. Identify — Classify the task into domain(s)
  3. Load — Read only relevant reference(s)
  4. Plan — Outline approach before executing
  5. Execute — Use the best tool for the job (CLI > API > manual)
  6. Verify — Check output quality before delivering
  7. Log — Record to mistakes.json if error, patterns.json if reusable, daily log always
  8. Report — Summarize what was done and what's next

Installed Skills Reference

These skills are already available and should be used directly when applicable. The omni skill supplements them with domain knowledge, not replaces them:

Communication & Messaging

  • discord — Discord operations
  • slack — Slack control (reactions, pins)
  • bluebubbles — iMessage via BlueBubbles
  • imsg — iMessage/SMS CLI
  • wacli — WhatsApp messaging
  • voice-call — Voice calls
  • xurl — X (Twitter) API

Productivity & Notes

  • apple-notes — macOS Notes via memo
  • apple-reminders — macOS Reminders via remindctl
  • bear-notes — Bear notes via grizzly
  • things-mac — Things 3 task manager
  • notion — Notion API (pages, databases, blocks)
  • obsidian — Obsidian vault management
  • trello — Trello boards/lists/cards
  • gog — Google Workspace (Gmail, Calendar, Drive, Contacts, Sheets, Docs)
  • himalaya — Email via IMAP/SMTP

Development & GitHub

  • github — GitHub CLI (issues, PRs, CI, review)
  • gh-issues — GitHub issues → auto-fix PRs
  • coding-agent — Delegate to Codex/Claude Code
  • mcporter — MCP server management
  • gemini — Gemini CLI for Q&A
  • oracle — Oracle CLI (prompt bundling, sessions)

Media & Creative

  • openai-image-gen — Batch image generation
  • nano-banana-pro — Gemini image gen/edit
  • openai-whisper — Local speech-to-text
  • openai-whisper-api — Cloud speech-to-text
  • sag — ElevenLabs TTS
  • sherpa-onnx-tts — Offline TTS
  • gifgrep — GIF search/download
  • video-frames — Video frame extraction
  • peekaboo — macOS UI capture & automation
  • songsee — Audio spectrograms
  • nano-pdf — PDF editing with NL

Music & Audio

  • spotify-player — Spotify playback/search
  • sonoscli — Sonos speaker control
  • blucli — BluOS audio
  • songsee — Audio visualization

Smart Home & IoT

  • openhue — Philips Hue lights
  • eightctl — Eight Sleep pods
  • camsnap — RTSP/ONVIF cameras
  • sonoscli — Sonos speakers

Web & Search

  • weather — Weather via wttr.in / Open-Meteo
  • goplaces — Google Places API
  • blogwatcher — RSS/Atom feed monitoring
  • summarize — URL/podcast/file summarization

System & Security

  • healthcheck — Security hardening & audits
  • 1password — 1Password CLI secrets
  • node-connect — OpenClaw node diagnostics
  • session-logs — Session log analysis
  • tmux — tmux session control

Platform & Ecosystem

  • canvas — UI canvas presentation
  • clawhub — Skill marketplace
  • skill-creator — Skill development
  • model-usage — Model cost tracking

Food & Ordering

  • ordercli — Foodora order tracking

When NOT to Use This Skill

If a specialized installed skill clearly matches the task, use that skill directly. Omni is the fallback and the glue — it fills gaps and handles cross-domain work.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

91.06%
按下载量换算934

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 openclaw skills install omni-master 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills