Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Nero1688/claude-academic-skills --skill tej-data-wranglergit clone --depth 1 https://github.com/Nero1688/claude-academic-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nero1688/claude-academic-skills/tej-data-wrangler)<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/tej-data-wrangler"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/tej-data-wrangler/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/tej-data-wrangler"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/tej-data-wrangler.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00293 | $0.03167 |
| Opus 5 | $0.00147 | $0.01584 |
| Sonnet 5 | $0.00059 | $0.00633 |
| Haiku 4.5 | $0.00029 | $0.00317 |
Grade A, and why
tej-data-wrangler scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TEJ 資料清理師(Data Wrangler)
最高原則:零靜默修正(Zero Silent Correction)。 你絕不悄悄刪一列、改一個值、或平滑一個極端值。真實財務數據就算長得奇怪也可能是真的(如虧損公司 ROE 為 -300%)。凡是動手,都要在清理報告裡列出「動了什麼、為什麼、影響幾列」,並保留原始檔不覆蓋。
Step 1|載入與結構盤點(先看清楚再動手)
- 讀檔明示編碼(TEJ 中文欄名常是 Big5 或 UTF-8,讀錯會亂碼);記錄原始列數、欄數(這是後面對帳的基準)。
- 印出前幾列與各欄型別,辨識:識別欄(公司代號、公司名、年/季)、數值欄、字串欄、日期欄。
- 辨識 TEJ 特有陷阱:遺漏被寫成
-/--/N/A/空字串;千分位逗號讓數值被讀成字串;金融業與一般業列混雜。
Step 2|欄名標準化與格式解析
- 欄名標準化:中文+代碼混排欄名 → 一致的英文 snake_case(如「資產總額(千元)」→
total_assets),並保留一張「原欄名↔新欄名」對照表寫進報告(可追溯)。 - 日期/年季:把「2023Q1」「11201」(民國年月)「2023/03/31」統一成標準年、季或日期欄;民國↔西元換算要註明(+1911)。
- 數值清理:去千分位逗號、把
-類遺漏標記統一轉成真正的 NA(只轉標記,不轉真值)。
Step 3|遺漏值分析報告(分析、不擅自填補)
- 逐欄遺漏數與比例;標出遺漏最嚴重的欄。
- 判斷遺漏型態:是否系統性(如某揭露 2005 才開始、ESG 早年整片缺、金融業某欄結構性為空)——系統性遺漏是樣本選擇偏誤的伏筆,要提醒使用者,不是隨手填補的對象。
- 填補只給建議、不預設執行:列出選項(listwise 刪除/該變數不填只揭露/可辯護的插補)與各自代價,讓使用者決定;面板資料尤其不建議跨公司均值亂填。
Step 4|品質紅旗檢查(L-003 核心,逐項跑)
- 不可能值:負的總資產、比率超出合理界(如負債比>1 需查是否為淨值為負的真實案例而非錯誤)、年份超出資料範圍、公司代號位數異常。標記為
Needs Review,不自動刪。 - 同值重複紅旗(L-003 真實教訓):掃描是否有「不該相同的欄格出現一模一樣的值」(如兩家公司整列數字全等、或某欄大量重複同一值)——這常是複製貼上或合併錯誤,回頭查來源。
- 重複列:同一公司-年出現多列(TEJ 匯出偶發),標記並讓使用者確認保留哪筆。
Step 5|面板長寬轉換與縮尾(選用,需揭露)
- 長寬轉換:依分析需求 wide↔long(如各年欄變成 year 欄的長格式),轉換前後都印列數對帳。
- winsorize 縮尾(選用):財務比率常在 1%/99% 縮尾以抑制極端值影響。這是選項不是預設;若使用者要做,套用後必附標準揭露句(見 output_contract),並在報告記錄縮尾前後的分位數變化。縮尾是「壓極端」不是「刪資料」,比刪除更可辯護,但仍須揭露。
Step 6|處理前後對帳與人工抽核(交付前必做)
- 列數對帳:原始 N → 各步驟後 N,逐步列出差額並解釋每一列是為什麼消失的(遺漏刪除?去重?篩選金融業?)。差額對不上就停,不交付。
- 抽 3 筆人工核對:隨機抽 3 個公司-年,把清理後的值倒推回原始 RAW 檔對照,確認沒洗錯。報告裡列出這 3 筆的對照。
<output_contract> 交付雙件套:
件一:清理腳本(R 或 Python,先問使用者慣用哪個;預設 Python/pandas 或 R/tidyverse)
- 區塊順序:
# 0 讀檔+盤點 → # 1 欄名/格式標準化 → # 2 遺漏分析 → # 3 品質紅旗 → # 4 長寬/縮尾(選用)→ # 5 對帳+抽核 → # 6 存乾淨檔。 - 逐行中文註解;不覆蓋原始檔,輸出另存
*_clean.csv。 - 開頭再現性聲明:套件版本、
set.seed(抽核/縮尾若涉隨機)、輸入輸出檔名與編碼。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 107 lines · 293 tokens per session scan A 6a2e9e6453c5
tej-data-wrangler is a skill published in the GitHub repository Nero1688/claude-academic-skills (6 stars, last pushed 9d ago), licensed MIT. It adds 293 tokens to every session and 3,167 once invoked, about $0.0015 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
alterlab-deep-research
Runs a 13-agent deep research pipeline for rigorous academic work on any topic across 7 modes (full research, quick brief, paper review, lit-review, fact-check, Socratic guided research dialogue, and systematic review with optional meta-analysis), covering research-question formulation, Socratic mentoring, methodology…
alterlab-paper-writer
Drafts and revises academic papers through a 12-agent pipeline with hardened LaTeX output (apa7 document class, justified text, table column-width formula, centered bilingual abstracts, standardized font stack, PDF compiled from LaTeX), supporting IMRaD, literature review, theoretical, case study, policy brief, and…
alterlab-research-pipeline
Orchestrates the full academic research pipeline (research, write, integrity check, review, revise, re-review, re-revise, final integrity check, finalize), coordinating alterlab-deep-research, alterlab-paper-writer, and alterlab-paper-reviewer into a seamless 10-stage workflow with mandatory integrity verification…
alterlab-imaging-data-commons
Query and download public cancer imaging data from the NCI Imaging Data Commons (IDC) using the idc-index Python package, filtering by metadata, visualizing in-browser, and checking licenses, with no authentication required. Use when obtaining large-scale radiology (CT, MR, PET) or digital pathology DICOM datasets for…
alterlab-pyhealth
Develops, tests, and deploys clinical machine learning models with the PyHealth healthcare AI toolkit. Use when working with electronic health records (EHR), clinical prediction tasks (mortality, readmission, drug recommendation), medical coding systems (ICD, NDC, ATC), physiological signals (EEG, ECG), healthcare…
alterlab-paper-reviewer
Simulates a full multi-reviewer journal review PANEL — 5 personas (Editor-in-Chief + 3 peer reviewers + a Devil's Advocate) debate a manuscript and produce a consensus Editorial Decision (accept/minor/major/reject) plus a prioritized Revision Roadmap. Modes: full, re-review (verify revisions addressed prior comments)…