SkillAgentSearch skills...

data-doc

Document datasets, variables, sources, and merge keys for replication

Install / Use

npx skills add brycewang-stanford/Auto-Empirical-Research-Skills --skill data-doc

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

86/100

Supported Platforms

Universal

Our assessment of data-doc

data-doc scores 86/100 on our quality scale, 704th of 1,155 Content & Media skills we index.

Its SKILL.md is 4.0 KB long, well organised into 11 sections with 2 code examples: a solid amount of guidance for an agent.

With 4,360 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
26/30
Structure
18/20
Description
12/15
Adoption
15/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so data-doc is actively maintained.
  • Our last check on 2026-11-03 found the source still online.
  • No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
  • Its trust signals score 88/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-09-27. Automated pattern scan on 2026-09-27. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

data-doc compared with similar skills

All 4 of these similar skills score higher than data-doc; compare them before choosing.

SkillScoreStarsUpdatedFormat
data-doc (this skill)by brycewang-stanford864.4k54d agoSKILL.md
siyuanby siyuan-note10047.3ktodayMCP Server
x402by browser-use100121.0k3d agoSKILL.md
algorithmic-artby anthropics100177.9k55d agoSKILL.md
pptxby anthropics100177.9k55d agoSKILL.md

Frequently asked questions

How do I install data-doc?
Run npx skills add brycewang-stanford/Auto-Empirical-Research-Skills --skill data-doc. The install tabs above show the steps for each supported agent.
Which AI agents does data-doc work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is data-doc safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 88/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is data-doc still maintained?
The repository was last updated about 2 months ago, so data-doc is actively maintained.

name: data-doc description: Document datasets, variables, sources, and merge keys for replication

Data Documentation Assistant

Help document datasets systematically for replication packages, co-author handoffs, and future reference.

Dataset Documentation Template

When documenting a dataset, capture:

1. Dataset Overview

  • Name: (e.g., firm_year_panel.parquet)
  • Location: (relative path from project root)
  • Created by: (script that generates it)
  • Last updated:
  • Unit of observation: (firm-year, loan-quarter, etc.)
  • N observations:
  • Time coverage:

2. Source Data

| Source | Access | Raw File | Notes | |--------|--------|----------|-------| | WRDS Compustat | Subscription | raw/compustat_funda.csv | Annual fundamentals | | Revelio Labs | Licensed | raw/revelio_positions.parquet | Via BU server | | Hand-collected | Manual | raw/manual_coding.xlsx | See coding protocol |

3. Key Variables

| Variable | Type | Description | Source | Notes | |----------|------|-------------|--------|-------| | gvkey | str | Compustat firm identifier | Compustat | Primary key | | fyear | int | Fiscal year | Compustat | | | at | float | Total assets ($ millions) | Compustat | Winsorized 1/99 | | treated | int | =1 if treated firm | Constructed | See section 4 |

4. Variable Construction

For constructed/derived variables, document:

Variable: treated
Definition: =1 if firm received first PE investment in year t
Construction:
  1. Merge PitchBook deals to Compustat on EIN
  2. Keep first deal per firm
  3. Flag year of first investment
Script: 2a_construct_treatment.py, lines 45-78

5. Sample Filters

Document all filters applied:

| Filter | Observations Dropped | Remaining | |--------|---------------------|-----------| | Raw data | - | 150,000 | | Drop financials (SIC 6000-6999) | 25,000 | 125,000 | | Require non-missing assets | 5,000 | 120,000 | | Require 2+ years in panel | 10,000 | 110,000 |

6. Merge Keys

| Dataset A | Dataset B | Key(s) | Match Rate | Notes | |-----------|-----------|--------|------------|-------| | Compustat | CRSP | gvkey | 95% | Via CCM link table | | Compustat | PitchBook | EIN | 72% | Manual cleaning needed | | Revelio | Compustat | company_name | 68% | Fuzzy match, see script |


Quick Commands

  • "document this dataset" - Generate full template for a dataset
  • "variable list" - Create variable table only
  • "merge documentation" - Focus on merge keys and match rates
  • "sample flow" - Generate sample filter table
  • "codebook" - Formal codebook format for replication package

Codebook Format (for Replication Packages)

================================================================================
CODEBOOK: firm_year_panel.dta
Generated: 2026-01-15
================================================================================

IDENTIFICATION
  gvkey         Compustat permanent firm identifier
  fyear         Fiscal year

OUTCOME VARIABLES
  roa           Return on assets = ni/at (winsorized 1/99)
  investment    Capex/lagged assets = capx/L.at

TREATMENT VARIABLES
  post          =1 for years after treatment
  treated       =1 for firms ever treated
  treat_post    Interaction: treated × post (DiD coefficient)

CONTROLS
  size          Log total assets = ln(at)
  leverage      Book leverage = (dltt+dlc)/at
  mtb           Market-to-book = (prcc_f×csho)/ceq

FIXED EFFECTS
  ff48          Fama-French 48 industry classification
  state         State of incorporation
================================================================================

Data Provenance Checklist

Before finalizing a dataset, verify:

  • [ ] All source files documented with access instructions
  • [ ] Variable definitions are unambiguous
  • [ ] Sample filters documented with observation counts
  • [ ] Merge match rates reported
  • [ ] Winsorization/trimming documented
  • [ ] Missing value treatment documented
  • [ ] Script that creates dataset is identified
  • [ ] Date of creation recorded

Related Skills

View on GitHub
GitHub Stars4.4k
CategoryContent
Updated1mo ago
Forks527

Languages

Stata

Trust signals

88/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 medium