Agent skill

Report Composition

by larry-zy in larry-zy/cognida

综合报告生成技能,把多段取数(总览+趋势+分群等)整合成一份带洞察的结构化报告. An agent skill from larry-zy/cognida.

MITAuto-check passed

Install Report Composition

skills CLI
$ npx skills add larry-zy/cognida --skill report-composition -a claude-code

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

GitHub CLI
$ gh skill install larry-zy/cognida report-composition --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/larry-zy/cognida.git skills-src && mkdir -p .claude/skills && cp -r skills-src/services/cognida-go/skills/report-composition .claude/skills/report-composition && 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
report-composition
GitHub stars
182
Token cost
~397 tokens
SKILL.md length
44 words
Files
1
Skills in repo
9
Repo updated
First seen
Licence
MIT

At a glance

综合报告生成技能,把多段取数(总览+趋势+分群等)整合成一份带洞察的结构化报告. An agent skill from larry-zy/cognida.

  • SKILL.md covers 何时用 / 何时不用, 标准流程, 工具 and 编排(多主题务必扇出隔离), plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Report Composition is an agent skill from larry-zy/cognida. 综合报告生成技能,把多段取数(总览+趋势+分群等)整合成一份带洞察的结构化报告。

Its SKILL.md is about 400 tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: 不止是能问数据的助手——受治理、可审计、可度量的企业级「数据+知识」Agent 平台. The licence is MIT.

Example prompts

  • “/report-composition”

What it can do on your machine

Read from SKILL.md and the folder at commit bf76257. 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 (its code samples are sql and markdown).

    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

Report Composition loads about 397 tokens when it runs. Until then it costs about 15 tokens; SKILL.md has 44 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~15
When it runs · the whole SKILL.md, loaded when a task matches
~397

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 larry-zy/cognida at commit bf76257, republished under its MIT licence (© larry-zy). 44 words, ~397 tokens.

Download SKILL.mdSave it as .claude/skills/report-composition/SKILL.md (or your agent's skills folder).
name
report-composition
description
综合报告生成技能,把多段取数(总览+趋势+分群等)整合成一份带洞察的结构化报告。
when_to_use
当用户要“一份报告/综合分析/经营概览”,需要多个维度、多段查询并整合成趋势与洞察时使用;单个数字或单条查询不属于此技能。
category
data
tags
report, analytics, composition, insight
version
1.0.0
author
Cognida Team

综合报告生成

面向「多段取数 → 整合 → 洞察」的报告场景:先拆解报告需要的指标块,分别取数,再汇总成带趋势与结论的结构化报告。

何时用 / 何时不用

  • ✅ 用:要一份报告/经营概览,涉及多个指标 + 趋势 + 分群/对比,需要整合叙述。
  • ❌ 不用:
    • 只要一个数或一张明细 → text2sql-adhoc
    • 只取单个治理指标 → semantic-metric

标准流程

1. 拆解报告结构:核心指标总览 / 趋势 / 分群 / 对比
   ↓
2. 逐块取数:能走语义层的用 semantic_query,其余用 sql_execute
   ↓
3. 计算派生指标(客单价、复购率等)并按时间聚合出趋势
   ↓
4. data_analysis 整合各段数据 + 洞察,输出结构化报告

工具

  • semantic_query:按治理口径取核心指标(优先)。
  • sql_execute:取语义层未覆盖的明细/派生计算(如按客户分组统计复购)。
  • data_analysis:把多段结果与洞察整合为报告。
复购率示例(派生计算)
sql
SELECT
  SUM(CASE WHEN cnt >= 2 THEN 1 ELSE 0 END) * 1.0 / COUNT(*) AS repurchase_rate
FROM (SELECT customer_id, COUNT(*) cnt FROM orders GROUP BY customer_id) t;

编排(多主题务必扇出隔离)

  • 报告多主题、每主题各自展开取数+分析,上下文最重:用 delegate_parallel 并行委派 Insight(每主题一个), 每个 Insight 内部自行取数+分析、只回传结论摘要 + result_id,你汇总各主题结论。不要在主循环里逐一 展开各主题,否则中间数据会撑爆上下文。
  • 仅单一主题的轻量报告可直接按标准流程 data_analysis(analysis_type=report) 一次成形。
  • 每次委派携最小 scope(read);汇总时以各 Insight 回传的 result_id 承载明细,正文只放结论与口径。

输出结构

markdown
## 经营指标综合报告
### 核心指标总览   —— GMV / 订单数 / 客单价
### 趋势分析       —— 按月聚合的走势
### 分群 / 复购     —— 复购率、关键客群
### 关键洞察与建议
### 数据来源与口径

约束

  • 每段取数标注口径与来源;派生指标写明公式。
  • 优先复用语义层口径,保证报告内各指标一致、可对齐。
  • 报告结论需有数据支撑,不臆断趋势成因。

© larry-zy, MIT. 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 services/cognida-go/skills/report-composition of larry-zy/cognida.

Open the folder on GitHubat commit bf76257

Compare with similar skills

Report Composition 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.

Report Composition compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Report Composition this skilllarry-zy/cognida182—~397Automated safety check: PassMIT
Compositionshadcn-labs/framecn150—~1.1kAutomated safety check: PassMIT
Composition Patternssickn33/agentic-awesome-skills47k1 repos~742Automated safety check: PassMIT
Vercel Composition Patternssupabase/supabase111k58 repos~726Automated safety check: PassMIT
Team Composition Patternswshobson/agents40k—~2kAutomated safety check: PassMIT
Atmos Compositionscloudposse/atmos1.4k—~486Automated safety check: PassApache-2.0

Similar skills

  • Composition

    shadcn-labs/framecn

    Create video compositions with Editframe using HTML web components or React.

    150 GitHub stars~1.1k tokensUpdated 2 days ago
    Media & CreativeAuto-check passed
  • Composition Patterns

    sickn33/agentic-awesome-skills

    A skill your agent uses when working with composition-patterns tasks or workflows

    47k GitHub starsUsed in 1 repo~742 tokens
    DevelopmentAuto-check passed
  • Official

    React composition patterns that scale. An agent skill from supabase/supabase.

    111k GitHub starsUsed in 58 repos~726 tokens
    DevelopmentAuto-check passed
  • Design optimal agent team compositions with sizing heuristics, preset configurations, and agent type selection.

    40k GitHub stars~2k tokensUpdated 4 days ago
    DevelopmentAuto-check passed
  • Atmos Compositions

    cloudposse/atmos

    Atmos compositions: named service groupings, compositions.<name.services, components.container[].composition, composition validate, and service contract checks

    1.4k GitHub stars~486 tokensUpdated today
    DevOps & CloudAuto-check passed
  • Design optimal team structures, hiring plans, compensation strategies, and equity allocation for early-stage startups from pre-seed through Series A.

    40k GitHub stars~1.9k tokensUpdated 4 days ago
    Product & Project ManagementAuto-check passed

More from larry-zy/cognida

All 9 skills in this repo
  • Attribution Analysis

    larry-zy/cognida

    归因/根因分析技能,把指标的期间变化拆解到候选维度,定位驱动因子并给出置信与下钻建议. An agent skill from larry-zy/cognida.

    182 GitHub stars~524 tokensUpdated 1 mo ago
    Auto-check passed
  • Semantic Metric

    larry-zy/cognida

    语义指标取数技能,先查治理好的指标语义模型,再按统一口径取指标,避免对物理表猜测. An agent skill from larry-zy/cognida.

    182 GitHub stars~329 tokensUpdated 1 mo ago
    Auto-check passed
  • Generate a comprehensive report of core e-commerce operating metrics (GMV, orders, average order value, repurchase)

    182 GitHub stars~470 tokensUpdated 1 mo ago
    Auto-check passed
  • Graph Retrieval

    larry-zy/cognida

    图谱与血缘检索技能,查询实体之间的关系、上下游依赖与数据血缘。

    182 GitHub stars~236 tokensUpdated 1 mo ago
    Auto-check passed
  • Text2sql Adhoc

    larry-zy/cognida

    即席取数技能,把一句话的具体数据问题翻译成 SQL 并执行返回结果。

    182 GitHub stars~390 tokensUpdated 1 mo ago
    Auto-check passed
  • Trend Analysis

    larry-zy/cognida

    趋势分析技能,取带时间维度的序列并量化其增长/变化幅度与异常点。

    182 GitHub stars~428 tokensUpdated 1 mo ago
    Auto-check passed

Questions about Report Composition

What does Report Composition do?

综合报告生成技能,把多段取数(总览+趋势+分群等)整合成一份带洞察的结构化报告. An agent skill from larry-zy/cognida. Report Composition is an agent skill from larry-zy/cognida.

How do I install Report Composition in Claude Code?

Run `npx skills add larry-zy/cognida --skill report-composition -a claude-code`. Or copy the skill folder (services/cognida-go/skills/report-composition in larry-zy/cognida) into .claude/skills/report-composition in your project. Claude Code loads it when a task matches its description.

How do I install Report Composition in Codex?

Run `npx skills add larry-zy/cognida --skill report-composition -a codex`. Or copy the skill folder (services/cognida-go/skills/report-composition in larry-zy/cognida) into .agents/skills/report-composition in your project. Codex loads it when a task matches its description.

Can I use Report Composition 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 larry-zy/cognida --skill report-composition -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/report-composition, .gemini/skills/report-composition, .github/skills/report-composition and .opencode/skills/report-composition in your project.

What does Report Composition need to run?

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

Does Report Composition 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 Report Composition 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 Report Composition use?

Report Composition is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Report Composition use?

About 397 tokens (SKILL.md is roughly 1.6k 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 Report Composition?

Skills that share tags, products or a category with Report Composition: Composition (shadcn-labs/framecn, 150 stars), Composition Patterns (sickn33/agentic-awesome-skills, 47k stars), Vercel Composition Patterns (supabase/supabase, 111k stars) and Team Composition Patterns (wshobson/agents, 40k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Report Composition?

larry-zy (a GitHub user) maintains it in larry-zy/cognida, which has 182 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on August 31, 2026.

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