data_analysis

仓库创建 2026年7月6日最近提交 21 天前SkillHot 收录 20 天前
▸ 精选理由

谨慎不编造数字,适合快速数据核查与展示

这个 Skill 做什么

解析用户提供的表格或日志并给出结论与证据表

把你贴过来的数据片段(CSV、表格、日志、数字)快速读懂并给出结论与证据表,先说结论再展示支持的数据。适合需要从杂乱数据里抽结论、做验证或找异常的场景。特点是不编造数字,会明确指出数据不足之处,并给出最多精简的证据表来支撑结论。

▸ 展开 SKILL.md 英文原文

Summarize, structure and interrogate data the user pastes or points to (CSV snippets, tables, logs, numbers).

数据与抓取数据分析CSV解析证据优先通用
0
Stars
0
Forks
5
仓库内 Skill
+0
7 日增星
安装 / 使用
给你的 Agent 一句话(通用)
帮我安装这个 skill:https://raw.githubusercontent.com/jerryhsieh991-lang/jarvis-os/main/skills/data_analysis/SKILL.md
或 curl 直取 SKILL.md
curl -fsSL "https://raw.githubusercontent.com/jerryhsieh991-lang/jarvis-os/main/skills/data_analysis/SKILL.md"
SKILL.MD 节选查看完整文件 ↗
# Role
You are a pragmatic data analyst. You never invent numbers. If the data given
is insufficient, you say exactly what's missing. You lead with the answer,
then show the supporting table.

# Inputs
- Pasted text/CSV/log lines, or a file path inside the vault.
- A question about that data.

# Outputs
Markdown:
```
## Answer (one sentence)
## Evidence (small table, max 10 rows)
## Caveats
```

# Tools
- The local LLM only. For anything heavier the user is told to run a local
  script — this skill never calls paid APIs.
- Findings are saved to `vault/10_Reports/` tagged [[Data]] plus a topic link.
via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有