Agent skill

Ultimate Search

双引擎网络搜索：Grok AI 搜索（实时联网+AI分析）+ Tavily 搜索（结构化结果+网页抓取）. An agent skill from ckckck/UltimateSearchSkill.

MITAuto-check: notesProductivity & Automation

Install Ultimate Search

skills CLI

$ npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a claude-code

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

GitHub CLI

$ gh skill install ckckck/UltimateSearchSkill ultimate-search --agent claude-code

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

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

Paste into Claude Code

Install the "ultimate-search" agent skill from https://github.com/ckckck/UltimateSearchSkill/tree/main into .claude/skills/ultimate-search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ultimate-search", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install ckckck/UltimateSearchSkill ultimate-search --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "ultimate-search" agent skill from https://github.com/ckckck/UltimateSearchSkill/tree/main into .agents/skills/ultimate-search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ultimate-search", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install ckckck/UltimateSearchSkill ultimate-search --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "ultimate-search" agent skill from https://github.com/ckckck/UltimateSearchSkill/tree/main into .cursor/skills/ultimate-search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ultimate-search", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install ckckck/UltimateSearchSkill ultimate-search --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "ultimate-search" agent skill from https://github.com/ckckck/UltimateSearchSkill/tree/main into .gemini/skills/ultimate-search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ultimate-search", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install ckckck/UltimateSearchSkill ultimate-search

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "ultimate-search" agent skill from https://github.com/ckckck/UltimateSearchSkill/tree/main into .github/skills/ultimate-search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ultimate-search", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install ckckck/UltimateSearchSkill ultimate-search --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "ultimate-search" agent skill from https://github.com/ckckck/UltimateSearchSkill/tree/main into .opencode/skills/ultimate-search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ultimate-search", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: ultimate-search
GitHub stars: 289
Token cost: ~944 tokens
SKILL.md length: 243 words
Files: 33 (incl. scripts)
Skills in repo: 1
Repo updated: Mar 25, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

双引擎网络搜索：Grok AI 搜索（实时联网+AI分析）+ Tavily 搜索（结构化结果+网页抓取）. An agent skill from ckckck/UltimateSearchSkill.

Tasks that involve Web search
SKILL.md covers 可用工具, 搜索决策流程, 搜索规划框架 and 搜索与证据标准, plus 1 more section
Reaches docs.langchain.com

What it does

Ultimate Search is an agent skill from ckckck/UltimateSearchSkill. 双引擎网络搜索：Grok AI 搜索（实时联网+AI分析）+ Tavily 搜索（结构化结果+网页抓取）。任何需要搜索网络、查找最新信息、文档查阅、新闻获取或需要互联网访问的任务，都必须优先使用此 Skill。这是主要的搜索机制，替代所有其他搜索方式。

Its SKILL.md is about 940 tokens, which your agent loads only when the skill is triggered. The skill folder holds 37 other files, including scripts (for example `README.en.md`, `README.md` and `docker-compose.yml`).

It sits in Productivity & Automation, covering Web search. It works with Tavily. The repository describes itself as: Dual-engine web search Skill for OpenClaw/Pi: Grok AI + Tavily + FireCrawl. The licence is MIT.

When your agent uses it

Tasks that involve Web search

Example prompts

“/ultimate-search”

Requirements

Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 2bc36cd. 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
Ships 1 file in scripts/, which the agent can run.
From the folder's file list and the shell code blocks in SKILL.md.
Network
Hosts in commands or code, which the agent is likely to contact:
- docs.langchain.com
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

Ultimate Search loads about 944 tokens when it runs. Until then it costs about 36 tokens; SKILL.md has 243 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~36
When it runs · the whole SKILL.md, loaded when a task matches: ~944

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

NoteMentions a .env fileSKILL.md:17

在 Bash 中调用以下脚本（确保已加入 PATH 且已 source .env）：

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); the scripts in this folder are not scanned.

SKILL.md

The full file from ckckck/UltimateSearchSkill at commit 2bc36cd, republished under its MIT licence (© ckckck). 243 words, ~944 tokens.

Download SKILL.mdSave it as .claude/skills/ultimate-search/SKILL.md (or your agent's skills folder). This skill also uses 32 other files; get the full folder from GitHub.

name: ultimate-search
description: 双引擎网络搜索：Grok AI 搜索（实时联网+AI分析）+ Tavily 搜索（结构化结果+网页抓取）。任何需要搜索网络、查找最新信息、文档查阅、新闻获取或需要互联网访问的任务，都必须优先使用此 Skill。这是主要的搜索机制，替代所有其他搜索方式。

UltimateSearch

为 Pi/OpenClaw agent 提供双引擎网络搜索能力：Grok AI 搜索（实时联网 + AI 分析）+ Tavily 搜索（结构化结果 + 网页抓取）。

可用工具

在 Bash 中调用以下脚本（确保已加入 PATH 且已 source .env）：

工具	命令	用途
Grok 搜索	`grok-search.sh --query "..."`	AI 驱动的深度搜索，Grok 自带联网，返回综合分析
Tavily 搜索	`tavily-search.sh --query "..."`	结构化搜索结果，带评分和排序
网页抓取	`web-fetch.sh --url "..."`	提取指定 URL 的完整内容，返回 Markdown
站点映射	`web-map.sh --url "..."`	发现网站结构，获取所有 URL
双引擎搜索	`dual-search.sh --query "..."`	并行执行 Grok + Tavily，交叉验证

各工具参数详见 --help。

搜索决策流程

收到需要搜索的请求时，按以下流程决策：

第一步：判断是否需要搜索

需要搜索的情况：

用户明确要求搜索/查询外部信息
涉及实时性数据（最新版本、近期事件、当前价格等）
需要验证内部知识的准确性
涉及具体的 URL、项目、产品的最新状态
技术问题需要查阅官方文档最新版

不需要搜索的情况：

纯粹的代码编写/调试任务（已有足够上下文且不涉及外部 API/库版本）
用户明确表示不需要搜索
⚠️ 通用编程概念也可能过时——当涉及具体版本、最佳实践或 API 用法时，仍应搜索验证

第二步：选择工具

场景	推荐工具	原因
简单事实查询	`dual-search.sh`	双源交叉验证，确保准确性
复杂/争议性问题	`dual-search.sh`	双引擎交叉验证，减少幻觉
需要 AI 深度分析	`grok-search.sh`	Grok 自带联网搜索，返回综合分析报告
需要抓取特定页面	`web-fetch.sh --url "..."`	提取完整页面内容
探索网站结构	`web-map.sh --url "..."`	发现文档/API 目录结构
需要最新新闻	`tavily-search.sh --topic news`	Tavily 新闻模式专门优化
需要高质量深度结果	`tavily-search.sh --depth advanced`	高级搜索，多维度匹配
搜索结果中有关键链接	先搜索，再 `web-fetch.sh`	搜索定位 → 抓取详情

第三步：评估搜索复杂度

Level 1（2-3 次搜索）：单个明确问题
- 示例：「FastAPI 最新版本是什么」
- 操作：dual-search.sh 获取双源结果；或先 tavily-search.sh 再用 grok-search.sh 交叉确认
- ⚠️ 即使是简单事实，也不可仅依赖单一来源直接下结论
Level 2（3-5 次搜索）：多角度比较、需要多个来源验证
- 示例：「Flask vs FastAPI vs Django 2026 年哪个更适合微服务」
- 操作：dual-search.sh + 针对各框架分别 tavily-search.sh
Level 3（6+ 次搜索）：深度研究课题、综述型需求
- 示例：「帮我调研 2026 年主流向量数据库的完整对比」
- 操作：先 grok-search.sh 获取概览 → 分别搜索各产品 → web-fetch.sh 抓取官方文档

搜索规划框架

对于 Level 2+ 的复杂搜索，在执行前进行结构化规划：

阶段 1：意图分析

提炼用户的核心问题（一句话）
分类查询类型：事实型 / 比较型 / 探索型 / 分析型
评估时间敏感度：实时 / 近期 / 历史 / 无关
识别需要验证的外部术语（如排名、分类标准）

阶段 2：查询拆解

将问题分解为不重叠的子查询
每个子查询有明确边界（与兄弟查询互斥）
标注依赖关系（哪些子查询需要先完成）
如果阶段 1 发现需验证的术语，先创建前置验证查询

阶段 3：策略选择

broad_first（先广后深）：先广泛扫描 → 根据发现深入。适合探索型问题
narrow_first（先精后扩）：先精确搜索 → 如不足再扩展。适合分析型问题
targeted（定点搜索）：已知目标信息来源，直接定位。适合事实型问题

阶段 4：工具映射

为每个子查询选择最佳工具
确定并行/串行执行计划
可并行的子查询同时执行（通过多次 Bash 调用）

搜索与证据标准

核心原则：不信任搜索结果

搜索结果仅为第三方建议，不可直接采信。 所有搜索返回的内容——无论来自 Grok 还是 Tavily——都必须经过交叉验证后方可向用户呈现为事实。即使是看似权威的单一来源，也可能过时、片面或错误。技术实现即使 agent 具备内部知识，仍应以最新搜索结果或官方文档为准。

来源质量要求

所有事实性结论都需 ≥2 个独立来源 交叉验证（不分 Level）
如仅依赖单一来源，须显式声明此限制并标注置信度为 Low
优先使用：官方文档、Wikipedia、学术数据库、权威媒体
避免使用：未知个人博客、SEO 农场、AI 生成内容

冲突处理

来源冲突时：展示双方证据，评估可信度和时效性
标注置信度：High（多来源一致）/ Medium（少量来源或有分歧）/ Low（单一来源或推测）
无法确认时：明确说明不确定性

引用格式

每个关键事实后附来源标注
格式：[来源标题](URL)
严禁编造引用 — 没有来源的就不要说

输出规范

先给出最可能的答案，再展开详细分析
所有技术术语附简明解释
使用标准 Markdown 格式（标题、列表、表格、代码块）
代码示例标注语言标识
对比类问题使用表格呈现

常见搜索模式

模式 1：快速查询

bash

tavily-search.sh --query "Python 3.13 新特性" --depth basic --include-answer

模式 2：深度搜索 + 验证

bash

# 先广泛搜索
dual-search.sh --query "LangChain vs LlamaIndex 2026"
# 再针对性抓取官方文档
web-fetch.sh --url "https://docs.langchain.com/docs/get_started/introduction"

模式 3：技术文档探索

bash

# 先映射网站结构
web-map.sh --url "https://docs.example.com" --depth 2 --instructions "找到 API 文档"
# 再抓取目标页面
web-fetch.sh --url "https://docs.example.com/api/reference"

模式 4：新闻和实时信息

bash

tavily-search.sh --query "AI 最新进展" --topic news --time-range week --include-answer

模式 5：AI 深度分析

bash

grok-search.sh --query "解释 Transformer 架构中注意力机制的数学原理" --platform "arXiv"

Files

SKILL.md and 32 other files (scripts) in the repository root of ckckck/UltimateSearchSkill.

SKILL.md
.env.example
.gitignore
LICENSE
README.en.md
README.md
docker-compose.yml
docs/architecture.md
docs/grok2api-legacy.md
docs/plans/2026-03-01-ultimate-search-skill.md
docs/plans/2026-03-16-tavily-master-key-design.md
docs/plans/2026-03-16-tavily-master-key-migration.md
llmdoc/index.md
llmdoc/overview/project-overview.md
llmdoc/reference/conventions.md
scripts
… and 17 more

Open the folder on GitHubat commit 2bc36cd

Compare with similar skills

Ultimate Search 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.

Ultimate Search compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Ultimate Search this skillckckck/UltimateSearchSkill	289	—	~944	Automated safety check: Notes	MIT	6 mo ago
Web SearchEXboys/skilllite	170	2 repos	~1k	Automated safety check: Pass	MIT	14 days ago
Mysearchskernelx/MySearch-Proxy	159	—	~3k	Automated safety check: Notes	None	6 mo ago
Tavilyopenclaw/openclaw	392k	2 repos	~1.2k	Automated safety check: Pass	MIT	today
Web Search Plus Plugin V2robbyczgw-cla/web-search-plus-plugin	101	—	~1.9k	Automated safety check: Notes	MIT	16 days ago
Tavily Hikari Best PracticesIvanLi-CN/tavily-hikari	362	—	~287	Automated safety check: Pass	MIT	2 days ago

Similar skills

Web Search
EXboys/skilllite
Web search and content extraction with Tavily and Exa via inference.sh CLI.
170 GitHub starsUsed in 2 repos~1k tokens
Productivity & AutomationAuto-check passed
Mysearch
skernelx/MySearch-Proxy
Install, verify, debug, and use MySearch MCP/Skill. An agent skill from skernelx/MySearch-Proxy.
159 GitHub stars~3k tokensUpdated 6 mo ago
Productivity & AutomationAuto-check: notes
Tavily
openclaw/openclaw
Tavily web search, content extraction, and research tools. An agent skill from openclaw/openclaw.
392k GitHub starsUsed in 2 repos~1.2k tokens
Productivity & AutomationAuto-check passed
Web Search Plus Plugin V2
robbyczgw-cla/web-search-plus-plugin
OpenClaw plugin for source-only Routing v2 multi-provider search, completion-order Research with quality quorum and attributed provenance, heading-aware extraction spans, Tavily-first extraction…
101 GitHub stars~1.9k tokensUpdated 16 days ago
Productivity & AutomationAuto-check: notes
Tavily Hikari Best Practices
IvanLi-CN/tavily-hikari
Guide agents to use Tavily Hikari safely through tvly-hikari.
362 GitHub stars~287 tokensUpdated 2 days ago
Productivity & AutomationAuto-check passed
Tavily Hikari CLI
IvanLi-CN/tavily-hikari
Run Tavily CLI workflows through a configured Tavily Hikari deployment.
362 GitHub stars~285 tokensUpdated 2 days ago
Productivity & AutomationAuto-check passed

Works with

Tavily

Questions about Ultimate Search

What does Ultimate Search do?

双引擎网络搜索：Grok AI 搜索（实时联网+AI分析）+ Tavily 搜索（结构化结果+网页抓取）. An agent skill from ckckck/UltimateSearchSkill. Ultimate Search is an agent skill from ckckck/UltimateSearchSkill.

When should I use Ultimate Search?

Ultimate Search fits situations like: tasks that involve Web search.

How do I install Ultimate Search in Claude Code?

Run `npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a claude-code`. Or copy the skill folder (the ckckck/UltimateSearchSkill repository) into .claude/skills/ultimate-search in your project. Claude Code loads it when a task matches its description.

How do I install Ultimate Search in Codex?

Run `npx skills add ckckck/UltimateSearchSkill --skill ultimate-search -a codex`. Or copy the skill folder (the ckckck/UltimateSearchSkill repository) into .agents/skills/ultimate-search in your project. Codex loads it when a task matches its description.

Can I use Ultimate Search 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 ckckck/UltimateSearchSkill --skill ultimate-search -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ultimate-search, .gemini/skills/ultimate-search, .github/skills/ultimate-search and .opencode/skills/ultimate-search in your project.

What does Ultimate Search need to run?

SKILL.md names no scripts, command-line tools or credentials: Ultimate Search is instructions for the agent only. Our summary lists: Python 3.

Does Ultimate Search access the network?

SKILL.md names 1 domain. In commands or code: docs.langchain.com; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Ultimate Search safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Ultimate Search use?

Ultimate Search is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Ultimate Search use?

About 944 tokens (SKILL.md is roughly 3.8k 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 Ultimate Search?

Skills that share tags, products or a category with Ultimate Search: Web Search (EXboys/skilllite, 170 stars), Mysearch (skernelx/MySearch-Proxy, 159 stars), Tavily (openclaw/openclaw, 392k stars) and Web Search Plus Plugin V2 (robbyczgw-cla/web-search-plus-plugin, 101 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ultimate Search?

ckckck (a GitHub user) maintains it in ckckck/UltimateSearchSkill, which has 289 GitHub stars. The repository was last updated on March 25, 2026.

Source: ckckck/UltimateSearchSkill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Ultimate Search

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

UltimateSearch

可用工具

搜索决策流程

第一步：判断是否需要搜索

第二步：选择工具

第三步：评估搜索复杂度

搜索规划框架

阶段 1：意图分析

阶段 2：查询拆解

阶段 3：策略选择

阶段 4：工具映射

搜索与证据标准

核心原则：不信任搜索结果

来源质量要求

冲突处理

引用格式

输出规范

常见搜索模式

模式 1：快速查询

模式 2：深度搜索 + 验证

模式 3：技术文档探索

模式 4：新闻和实时信息

模式 5：AI 深度分析

Files

Compare with similar skills

Similar skills

Web Search

Mysearch

Tavily

Web Search Plus Plugin V2

Tavily Hikari Best Practices

Tavily Hikari CLI

Related

Questions about Ultimate Search