Agent skill

Skill Hub Guide

by agentscope-ai in agentscope-ai/QwenPaw-Data

Data Skill-Hub 技能仓库使用指南。在执行任何数据分析任务前,必须优先加载此技能以了解可用技能体系和调用规范。

Apache-2.0Auto-check passedData & Analytics

Install Skill Hub Guide

skills CLI
$ npx skills add agentscope-ai/QwenPaw-Data --skill skill-hub-guide -a claude-code

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

GitHub CLI
$ gh skill install agentscope-ai/QwenPaw-Data skill-hub-guide --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/agentscope-ai/QwenPaw-Data.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/qwenpaw-data-skills/skills/meta/skill-hub-guide .claude/skills/skill-hub-guide && 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
skill-hub-guide
GitHub stars
124
Token cost
~1.6k tokens
SKILL.md length
135 words
Files
1
Skills in repo
29
Repo updated
First seen
Licence
Apache-2.0

At a glance

Data Skill-Hub 技能仓库使用指南。在执行任何数据分析任务前,必须优先加载此技能以了解可用技能体系和调用规范。

  • Works in 6 steps: BI 业务分析链路(任务类型 = 2a) → 数据探索链路(任务类型 = 2b) → 统计建模链路(任务类型 = 2c) → …
  • Data & Analytics work in your project
  • SKILL.md covers 核心原则, 技能体系 and 强制约束
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Skill Hub Guide is an agent skill from agentscope-ai/QwenPaw-Data. Data Skill-Hub 技能仓库使用指南。在执行任何数据分析任务前,必须优先加载此技能以了解可用技能体系和调用规范。

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Data & Analytics. The repository describes itself as: Agentic enterprise data analytics: governed facts (DataBridge), reusable methodology (Skill-Hub), and controllable execution (Host). The licence is Apache-2.0.

When your agent uses it

  • Data & Analytics work in your project

Example prompts

  • “/skill-hub-guide”

Workflow steps

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

  1. BI 业务分析链路(任务类型 = 2a)
  2. 数据探索链路(任务类型 = 2b)
  3. 统计建模链路(任务类型 = 2c)
  4. 定量计算链路(任务类型 = 2d)
  5. 报告生成链路(任务类型 = 2e)
  6. 数据查询链路(任务类型 = 1a / 1b)

What it can do on your machine

Read from SKILL.md and the folder at commit e0bae36. 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

    No scripts in the folder and no shell commands in SKILL.md.

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

  • Network

    No URLs in SKILL.md.

    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

Skill Hub Guide loads about 1.6k tokens when it runs. Until then it costs about 19 tokens; SKILL.md has 135 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~19
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

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 agentscope-ai/QwenPaw-Data at commit e0bae36, republished under its Apache-2.0 licence (© agentscope-ai). 135 words, ~1,583 tokens.

Download SKILL.mdSave it as .claude/skills/skill-hub-guide/SKILL.md (or your agent's skills folder).
name
skill-hub-guide
description
Data Skill-Hub 技能仓库使用指南。在执行任何数据分析任务前,必须优先加载此技能以了解可用技能体系和调用规范。

skill-hub-guide

Data Skill-Hub 是由资深数据分析专家团队设计、经过大量真实业务场景验证的通用技能仓库。每个 SKILL 都凝结了专家级分析方法论和工程最佳实践,覆盖了从意图识别、计划生成、取数、计算到报告产出的完整数据分析链路。

严格遵循本仓库中的 SKILL 定义,你将获得专家级的分析质量——逻辑严谨、结论有据、产物规范。偏离 SKILL 定义则会导致分析遗漏、结论失准或产物不可用。

你的所有数据分析行为必须由 SKILL 驱动,严格遵循各 SKILL.md 中定义的执行步骤和规则,不得自行发挥。


核心原则

  1. SKILL 即规范:每个 SKILL.md 由领域专家编写,定义了该能力唯一正确的执行流程。即使你认为可以简化,也必须完整遵循
  2. 脚本优先:有脚本(scripts/)必须调脚本,不可用时才降级为自行实现
  3. 参数优先级:用户显式指定 > 域知识包 > 语义层接口 > SKILL 默认值
  4. 按需调用:禁止一次性预读所有 SKILL,根据当前执行到的子任务和步骤逐个加载对应 SKILL

技能体系

分层架构

技能按所在目录划分为 6 层。下表中「目录」一栏即 skills/ 下的物理目录,新增 SKILL 时按其能力归属放入对应目录,并在本表登记。

层级目录职责SKILL
L0 路由skills/routing/判定用户意图属于哪种任务类型(查询/分析/建模/报告/非数据),将请求分发至对应处理链路data-intent-router
L1 规划skills/planning/补充上下文、匹配分析模块和指标,将用户需求转化为结构化的可执行分析计划analysis-plan-builder
Workflowsskills/workflows/按分析计划编排原子技能与取数能力,完成从数据获取到结论汇总的完整分析流程fetch-data(取数)<br>bi-metric-analysis(指标观测与异常归因)<br>bi-retention-analysis(留存率分析)<br>bi-conversion-analysis(转化率分析)<br>bi-cohort-analysis(同期群分析)
Atomicskills/atomic/执行单一、独立的分析动作(计算、检测、下拆、归因、聚类、报告等)指标观测与计算:bi-metric-observation、bi-retention-rate、bi-conversion-rate<br>异常检测与阈值:bi-anomaly-detection、bi-adaptive-threshold<br>归因与下钻:bi-dimension-drilldown、bi-attribution-analysis、bi-time-impact-attribution、bi-new-dimension-analysis<br>统计与画像:bi-event-analysis、bi-funnel-analysis、bi-comparison-analysis、bi-distribution-analysis、bi-clustering、bi-ltv-analysis<br>报告输出:bi-report-generation
Runtimeskills/runtime/横切面强制规范,约束所有层的执行行为(产物落盘、数据复用、异常处理、语义层查询)runtime-guide、bi-semantic-layer-guide
Metaskills/meta/仓库使用与协作约定,加载其它任何 SKILL 之前必须先读skill-hub-guide(本文件)
典型调用链路

下面给出几种常见任务类型的链路。Runtime 层(runtime-guide、bi-semantic-layer-guide)在所有链路里全程生效,不再在每条链路中重复展开;fetch-data 在任何需要新取数的链路前置生效。

1. BI 业务分析链路(任务类型 = 2a)

由 data-intent-router 判定为业务分析后,由 analysis-plan-builder 给出结构化计划,再根据计划落到具体的 workflow。

1.1 通用指标分析(含异常归因)

用户问题
  → L0 data-intent-router          判定为 BI 业务分析(2a)
  → L1 analysis-plan-builder       生成结构化分析计划,确认指标与角色
  → Workflows fetch-data           按计划取数落盘
  → Workflows bi-metric-analysis   编排以下原子技能:
      → bi-metric-observation
          观测北极星指标表现,确认基准数据
      → bi-anomaly-detection
          对北极星指标做异常检测
          ├─ [无外部阈值] → bi-adaptive-threshold 自适应计算阈值
          └─ [有域知识包/用户指定阈值] → 直接使用
          ├─ [发现异常]
          │   → bi-dimension-drilldown
          │       按归因维度逐层下拆
          │       └─ bi-attribution-analysis 计算贡献度(定位哪个维度异常)
          │   → [可选] bi-causal-attribution
          │       从外部文档证据(周报/活动记录/产品发布)中解释"为什么"
          │       └─ [可选] bi-time-impact-attribution
          │               量化已找到的事件影响度
          │   → [无外部证据时] bi-time-impact-attribution
          │       直接拆解结构变动/趋势变动/事件影响
          │       └─ [需要阈值] → bi-adaptive-threshold
          └─ [未发现异常] 继续
      → bi-new-dimension-analysis
          识别新增维度值并评估表现
  → Atomic bi-report-generation    (如需输出报告)生成 HTML 报告

1.2 留存率分析

用户问题(涉及次日/第 N 日留存、访问/使用留存、留存差异等)
  → L0 data-intent-router          判定为 BI 业务分析(2a)—— 留存子类
  → L1 analysis-plan-builder       明确留存锚点、留存指标、对比范围
  → Workflows fetch-data           取数(day0 / dayn 用户数)
  → Workflows bi-retention-analysis 编排:
      → [可选] bi-clustering           按用户特征分群
      → bi-retention-rate              计算各群/整体的留存率
      → [可选] bi-comparison-analysis  跨时间/群体/留存类型对比
  → Atomic bi-report-generation    (如需输出报告)

1.3 转化率分析

用户问题(涉及转化率、阶段转化、转化变化归因等)
  → L0 data-intent-router          判定为 BI 业务分析(2a)—— 转化子类
  → L1 analysis-plan-builder       明确转化口径、起止行为、对比范围
  → Workflows fetch-data           取数(分子/分母用户数或事件数)
  → Workflows bi-conversion-analysis 编排:
      → [可选] bi-clustering           按用户/产品特征分群
      → bi-conversion-rate             计算各群/整体的转化率
      → [可选] bi-anomaly-detection    对转化率序列做异常检测
      → [可选] bi-comparison-analysis  跨时间/群体/渠道对比
  → Atomic bi-report-generation    (如需输出报告)

1.4 同期群(cohort)分析

用户问题(按起始特征分群后跟踪后续表现)
  → L0 data-intent-router          判定为 BI 业务分析(2a)—— cohort 子类
  → L1 analysis-plan-builder       明确起始特征、跟踪指标、观察窗口
  → Workflows fetch-data           取数(cohort 划分依据 + 后续期表现指标)
  → Workflows bi-cohort-analysis   编排:
      → bi-clustering                  按起始特征划分 cohort
      → 后续表现指标计算               按需调用 bi-retention-rate / bi-conversion-rate / bi-metric-observation 等
      → bi-comparison-analysis         跨 cohort 对比
  → Atomic bi-report-generation    (如需输出报告)
2. 数据探索链路(任务类型 = 2b)
用户问题
  → L0 data-intent-router          判定为数据探索(2b)
  → L1 analysis-plan-builder       数据探查 + 生成子任务序列
  → Workflows fetch-data           按子任务序列按需取数
  → 按子任务序列调用 Atomic:
      bi-distribution-analysis(分布、缺失、异常值)、
      bi-event-analysis(事件计数与聚合)、
      bi-clustering(密度型子结构发现)等
  → 无固定 Workflows 编排,由 agent 按计划串接
3. 统计建模链路(任务类型 = 2c)
用户问题(聚类、画像、假设检验等)
  → L0 data-intent-router          判定为统计建模(2c)
  → L1 analysis-plan-builder       明确建模目标、特征、评估口径
  → Workflows fetch-data           取建模所需样本数据
  → 按目标调用 Atomic:
      bi-clustering(聚类/象限/分层)、
      bi-comparison-analysis(显著性/差异检验)等
  → Atomic bi-report-generation    (如需输出报告)
4. 定量计算链路(任务类型 = 2d)
用户问题(已知公式的指标计算,如 LTV、漏斗、转化、留存、分布等)
  → L0 data-intent-router          判定为定量计算(2d)
  → Workflows fetch-data           取计算所需数据
  → 直接调用对应 Atomic:
      bi-ltv-analysis、bi-funnel-analysis、bi-conversion-rate、
      bi-retention-rate、bi-event-analysis、bi-distribution-analysis 之一或多个
  → 返回计算结果(一般无需走 L1 规划)
5. 报告生成链路(任务类型 = 2e)
上下文中已有分析结论
  → L0 data-intent-router          判定为报告生成(2e)
  → Atomic bi-report-generation    将既有结论组织为可视化 HTML 报告
6. 数据查询链路(任务类型 = 1a / 1b)
用户问题
  → L0 data-intent-router          判定为元数据查询(1a)或数据查询(1b)
  → Workflows fetch-data           直接取数或读元数据返回,不经过规划和分析流程

强制约束

必须做
  • 每个步骤必须查阅对应 SKILL.md,按其定义的步骤顺序执行
  • 产物必须落盘,遵循 runtime-guide 定义的目录和命名规范;取数产物遵循 fetch-data 定义的存储路径
  • 同一份数据只获取一次,不重复计算
  • 异常如实记录,不隐藏、不替代、不编造
  • 语义层可用时,按 bi-semantic-layer-guide 优先查询指标定义与维度信息,再进入业务计算
禁止做
  • 禁止跳过 SKILL 定义的步骤(如跳过异常检测直接归因、跳过取数直接编结果)
  • 禁止忽略脚本(脚本可用时不得自行实现替代)
  • 禁止编造数据(所有数字必须来自数据文件)
  • 禁止自行发明分析方法(SKILL 未定义的方法不得使用)

© agentscope-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in packages/qwenpaw-data-skills/skills/meta/skill-hub-guide of agentscope-ai/QwenPaw-Data.

Open the folder on GitHubat commit e0bae36

Compare with similar skills

Skill Hub Guide 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.

Skill Hub Guide compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Skill Hub Guide this skillagentscope-ai/QwenPaw-Data124—~1.6kAutomated safety check: PassApache-2.0
MatplotlibzLanqing/codex-claude-academic-skills4.7k17 repos~2.9kAutomated safety check: PassMIT
Exploratory Data Analysisspacering-net/codeg3.9k14 repos~3.6kAutomated safety check: PassMIT
Scikit LearnzLanqing/codex-claude-academic-skills4.7k16 repos~3.9kAutomated safety check: PassBSD-3-Clause
Chart Visualizationbytedance/deer-flow84k2 repos~840Automated safety check: PassMIT
TimesFM Forecastinggoogle-research/timesfm34k—~4.7kAutomated safety check: PassApache-2.0

Similar skills

  • Matplotlib

    zLanqing/codex-claude-academic-skills

    Low-level plotting library for full customization. An agent skill from zLanqing/codex-claude-academic-skills.

    4.7k GitHub starsUsed in 17 repos~2.9k tokens
    Data & AnalyticsAuto-check passed
  • Exploratory Data Analysis

    spacering-net/codeg

    Perform comprehensive exploratory data analysis on scientific data files across 200+ file formats.

    3.9k GitHub starsUsed in 14 repos~3.6k tokens
    Data & AnalyticsAuto-check passed
  • Scikit Learn

    zLanqing/codex-claude-academic-skills

    Machine learning in Python with scikit-learn. An agent skill from zLanqing/codex-claude-academic-skills.

    4.7k GitHub starsUsed in 16 repos~3.9k tokens
    Data & AnalyticsAuto-check passed
  • Chart Visualization

    bytedance/deer-flow

    Picks a suitable chart type from 26 options for your data, maps the data to that chart's parameters and generates a chart image through a JavaScript script.

    84k GitHub starsUsed in 2 repos~840 tokens
    Data & AnalyticsAuto-check passed
  • TimesFM Forecasting

    google-research/timesfm

    Forecasts any univariate time series zero-shot with Google's TimesFM model, returning point forecasts and calibrated prediction intervals without training.

    34k GitHub stars~4.7k tokensUpdated 10 days ago
    Data & AnalyticsAuto-check passed
  • Sandbox Bench

    vercel/next.js

    Official

    Benchmark React or Next.js changes on Vercel Sandbox VMs with paired A/B statistics: react PR/commit vs base, or Next.js PR/commit vs base, measured end-to-end through the bench/render-pipeline app…

    143k GitHub stars~4.1k tokensUpdated today
    Data & AnalyticsAuto-check passed

More from agentscope-ai/QwenPaw-Data

All 29 skills in this repo
  • Bi Report Generation

    agentscope-ai/QwenPaw-Data

    将 BI 数据分析结果组织成可视化 HTML 报告。当分析完成、需要生成报告时调用. An agent skill from agentscope-ai/QwenPaw-Data.

    124 GitHub stars~1.2k tokensUpdated 4 days ago
    Auto-check passed
  • Fetch Data

    agentscope-ai/QwenPaw-Data

    取数 / 查数据 / 拉数据 / 跑 SQL。把自然语言取数需求转为 SQL,经数据湖仓执行后返回查询结果供下游分析。任何需要业务数据的任务在工作区缺少对应文件时都必须先调用此技能——覆盖 BI 业务分析、留存 / 转化 / 同期群分析、数据探索 EDA、统计建模、定量计算、元数据查询、数据查询。命中任一即触发:(1) 直接索要指标或记录,如「DAU 多少」「上月销售额」「3…

    124 GitHub stars~2.5k tokensUpdated 4 days ago
    Auto-check passed
  • Bi Adaptive Threshold

    agentscope-ai/QwenPaw-Data

    通过量化历史数据的自然波动幅度,自适应计算判定阈值。当需要从数据本身确定阈值(如波动阈值、影响度阈值等)、而非使用固定值时调用。仅适用于日/周粒度阈值确定。

    124 GitHub stars~726 tokensUpdated 4 days ago
    Auto-check passed
  • Bi Anomaly Detection

    agentscope-ai/QwenPaw-Data

    基于阈值检测时间序列中的显著异常波动点。当需要找出指标异常波动日期、识别数据异动时调用. An agent skill from agentscope-ai/QwenPaw-Data.

    124 GitHub stars~637 tokensUpdated 4 days ago
    Auto-check passed
  • Bi Attribution Analysis

    agentscope-ai/QwenPaw-Data

    计算各维度(组)值对指标变动的贡献度,支持可加型量值指标和加权平均型/率值指标。当需要计算贡献度、解释指标"为什么涨/跌"时调用。

    124 GitHub stars~1.2k tokensUpdated 4 days ago
    Auto-check passed
  • Bi Causal Attribution

    agentscope-ai/QwenPaw-Data

    从运营周报、活动文档、对话输入或文档工具 API 中提取业务事件,与指标异常时间窗口对齐,生成有证据支撑的因果归因假设并排序。当已知指标存在异常波动、需要从外部文档证据中解释"为什么"时调用。

    124 GitHub stars~1.3k tokensUpdated 4 days ago
    Auto-check passed

Questions about Skill Hub Guide

What does Skill Hub Guide do?

Data Skill-Hub 技能仓库使用指南。在执行任何数据分析任务前,必须优先加载此技能以了解可用技能体系和调用规范。. Skill Hub Guide is an agent skill from agentscope-ai/QwenPaw-Data.

When should I use Skill Hub Guide?

Skill Hub Guide fits situations like: data & Analytics work in your project.

How do I install Skill Hub Guide in Claude Code?

Run `npx skills add agentscope-ai/QwenPaw-Data --skill skill-hub-guide -a claude-code`. Or copy the skill folder (packages/qwenpaw-data-skills/skills/meta/skill-hub-guide in agentscope-ai/QwenPaw-Data) into .claude/skills/skill-hub-guide in your project. Claude Code loads it when a task matches its description.

How do I install Skill Hub Guide in Codex?

Run `npx skills add agentscope-ai/QwenPaw-Data --skill skill-hub-guide -a codex`. Or copy the skill folder (packages/qwenpaw-data-skills/skills/meta/skill-hub-guide in agentscope-ai/QwenPaw-Data) into .agents/skills/skill-hub-guide in your project. Codex loads it when a task matches its description.

Can I use Skill Hub Guide 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 agentscope-ai/QwenPaw-Data --skill skill-hub-guide -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/skill-hub-guide, .gemini/skills/skill-hub-guide, .github/skills/skill-hub-guide and .opencode/skills/skill-hub-guide in your project.

What does Skill Hub Guide need to run?

SKILL.md names no scripts, command-line tools or credentials: Skill Hub Guide is instructions for the agent only.

Does Skill Hub Guide access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Skill Hub Guide 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 Skill Hub Guide use?

Skill Hub Guide is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Skill Hub Guide use?

About 1.6k tokens (SKILL.md is roughly 6.3k 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 Skill Hub Guide?

Skills that share tags, products or a category with Skill Hub Guide: Matplotlib (zLanqing/codex-claude-academic-skills, 4.7k stars), Exploratory Data Analysis (spacering-net/codeg, 3.9k stars), Scikit Learn (zLanqing/codex-claude-academic-skills, 4.7k stars) and Chart Visualization (bytedance/deer-flow, 84k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Skill Hub Guide?

agentscope-ai (a GitHub organization) maintains it in agentscope-ai/QwenPaw-Data, which has 124 GitHub stars. The repository holds 29 skills in this directory. The repository was last updated on October 5, 2026.

Source: agentscope-ai/QwenPaw-Data on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.