agentleFS
Barrios88Barrios88/barrios-skills/skills/econometrics/wrds/SKILL.md

wrds

Skill63 starsChanged 5 months agoMITView on GitHub

This skill should be used when the user asks to "query WRDS", "access Compustat", "get CRSP data", "pull Form 4 insider data", "query ISS compensation", "download SEC EDGAR filings", "get ExecuComp data", "access Capital IQ", "query TRACE bond data", "clean TRACE", "merge FISD", "query TAQ", "pull intraday trades or quotes", "build NBBO", "compute effective spread", or needs WRDS PostgreSQL query patterns.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 9 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Start the tunnel: bash mcp/wrds-mcp/tunnel/tunnel_up.sh and approve Duo MFA once per session

    line 18
  2. Very specific:

    When MCP tools are available (wrds_run_sql, wrds_get_compustat, etc.), prefer MCP over raw Python connections.

    line 23
  3. Distinctive:

    If the user has not configured MCP credentials, stop and walk them through install/wrds-mcp.md — do not guess usernames or passwords.

    line 25
  4. Distinctive:

    Before executing ANY WRDS query, you MUST:

    line 46
  5. Distinctive:

    Before EVERY query execution:

    line 75
  6. Distinctive:

    Always include for clean fundamental data:

    line 183
  7. Distinctive:

    Always use parameterized queries (never string formatting):

    line 248
  8. Distinctive:

    Use scalar parameter binding for single values:

    line 250
  9. Distinctive:

    Use ANY() for list parameters:

    line 257
  10. Distinctive:

    When querying historical data, leverage current date context for dynamic range calculations.

    line 299
  11. Distinctive:

    Always incorporate current date awareness in date-dependent queries to ensure results remain fresh across time.

    line 321

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Barrios Skills — John Barrios's curated workflow for economists and accountants. Prioritize reproducible empirical work, clear identification language, and journal-ready output.

WRDS MCP (recommended for AI agents)

John Barrios's wrds-mcp server exposes WRDS to Cursor, Claude, and other MCP clients.

Before querying WRDS, the user must:

  1. Install the MCP from mcp/wrds-mcp/ — see install/wrds-mcp.md
  2. Set their own credentials: WRDS_USERNAME and WRDS_PASSWORD (from their WRDS account)
  3. Start the tunnel: bash mcp/wrds-mcp/tunnel/tunnel_up.sh and approve Duo MFA once per session
  4. Register the MCP with WRDS_TUNNEL_PORT=49600 in Cursor or Claude config

When to use WRDS instead / with EDGAR: Compustat fundamentals panels, CRSP returns, CCM links, ExecuComp, ISS, TRACE/TAQ — use this skill + wrds-mcp. For live filing text, XBRL facts, and public Forms 3/4/5 without WRDS panel infrastructure, use the sec-edgar skill + SEC EDGAR MCP.

When MCP tools are available (wrds_run_sql, wrds_get_compustat, etc.), prefer MCP over raw Python connections. Still apply all filter rules in this skill.

If the user has not configured MCP credentials, stop and walk them through install/wrds-mcp.md — do not guess usernames or passwords.

Contents

  • WRDS MCP (recommended for AI agents)
  • Query Enforcement
  • Quick Reference: Table Names
  • Connection
  • Critical Filters
  • Parameterized Queries
  • Additional Resources

WRDS Data Access

WRDS (Wharton Research Data Services) provides academic research data via PostgreSQL at wrds-pgdata.wharton.upenn.edu:9737.

Query Enforcement

IRON LAW: NO QUERY WITHOUT FILTER VALIDATION FIRST

Before executing ANY WRDS query, you MUST:

  1. IDENTIFY what filters are required for this dataset
  2. VALIDATE the query includes those filters
  3. VERIFY parameterized queries (never string formatting)
  4. EXECUTE the query
  5. INSPECT a sample of results before claiming success

This is not negotiable. Claiming query success without sample inspection is LYING to the user about data quality.

Rationalization Table - STOP If You Think:

ExcuseRealityDo Instead
"I'll add filters later"You'll forget and pull bad dataAdd filters NOW, before execution
"User didn't specify filters"Standard filters are ALWAYS requiredApply Critical Filters section defaults
"Just a quick test query"Test queries with bad filters teach bad patternsUse production filters even for tests
"I'll let the user filter in pandas"Pulling millions of unnecessary rows wastes time/memoryFilter at database level FIRST
"The query worked, so it's correct"Query success ≠ data qualityINSPECT sample for invalid records
"I can use f-strings for simple queries"SQL injection risk + wrong type handlingALWAYS use parameterized queries

Red Flags - STOP Immediately If You Think:

  • "Let me run this query quickly to see what's there" → NO. Check Critical Filters section first.
  • "I'll just pull everything and filter later" → NO. Database-level filtering is mandatory.
  • "The table name is obvious from the request" → NO. Check Quick Reference section for exact names.
  • "I can inspect the data after the user sees it" → NO. Sample inspection BEFORE claiming success.

Query Validation Checklist

Before EVERY query execution:

For Compustat queries (comp.funda, comp.fundq):

  • Includes indfmt = 'INDL'
  • Includes datafmt = 'STD'
  • Includes popsrc = 'D'
  • Includes consol = 'C'
  • Uses parameterized queries for variables
  • Date range is explicitly specified

For CRSP v2 queries (crsp.dsf_v2, crsp.msf_v2):

  • Post-query filter: sharetype == 'NS'
  • Post-query filter: securitytype == 'EQTY'
  • Post-query filter: securitysubtype == 'COM'
  • Post-query filter: usincflg == 'Y'
  • Post-query filter: issuertype.isin(['ACOR', 'CORP'])
  • Uses parameterized queries

For Form 4 queries (tr_insiders.table1):

  • Transaction type filter specified (acqdisp)
  • Transaction codes specified (trancode)
  • Date range is explicitly specified
  • Uses parameterized queries

For TAQ queries (taqm_YYYY.ctm_ / cqm_ / complete_nbbo_*):**

  • Using taqm_YYYY (daily partitioned) for post-2003 work, not legacy taq
  • Trade filter: tr_corr = '00' (string, not integer)
  • Trade conditions: drop tr_scond containing Z, O, T, U, G, B (non-regular)
  • Regular trading hours: time_m BETWEEN '09:30:00' AND '16:00:00'
  • Price > 0 and size > 0 sanity checks
  • For NBBO: use complete_nbbo_* (not nbbom_* — incomplete by design)
  • If joining to CRSP: use wrdsapps_link_crsp_taqm with sdate/edate bounds
  • Consider wrds_iid_YYYY for pre-computed daily liquidity measures
  • Never SELECT * — specify columns (ctm has 20, cqm/nbbom are huge)
  • Sort explicitly by (time_m, time_m_nano, tr_seqnum) — insertion order not guaranteed

For TRACE queries (trace_enhanced.trace_enhanced):

  • Using trace_enhanced (uncapped) not trace_standard (capped) for research
  • Dick-Nielsen cleaning applied: drop trc_st IN ('X','C','W','Y') + matched originals
  • Pre-/post-2012-02-06 logic branch selected correctly
  • Agency double-count filter: NOT (cmsn_trd='Y' AND cntra_mp_id='D')
  • Commissioned trades handled (cmsn_trd='N' for price/yield work)
  • Special trades excluded: spcl_trd_fl != 'Y' (for price/liquidity analysis)
  • FISD merge for bond characteristics (coupon, maturity, rating, issuer)
  • Date range is explicitly specified on trd_exctn_dt
  • Uses parameterized queries

For ALL queries:

  • Sample inspection with .head() or .sample() BEFORE claiming success
  • Row count verification (is result size reasonable?)
  • NULL value check on critical columns
  • Date range validation (does min/max match expectations?)

Quick Reference: Table Names

DatasetSchemaKey Tables
Compustatcompcompany, funda, fundq, secd
ExecuCompcomp_execucompanncomp
CRSPcrspdsf, msf, stocknames, ccmxpf_lnkhist
CRSP v2crspdsf_v2, msf_v2, stocknames_v2
Form 4 Insiderstr_insiderstable1, header, company
ISS Incentive Labiss_incentive_labcomppeer, sumcomp, participantfy
Capital IQciqwrds_compensation
IBEStr_ibesdet_epsus, statsum_epsus
SEC EDGARwrdssecwrds_forms, wciklink_cusip
SEC Searchwrds_sec_searchfiling_view, registrant
EDGARedgarfilings, filing_docs
Fama-Frenchfffactors_monthly, factors_daily
LSEG/Datastreamtr_dsds2constmth, ds2indexlist
TRACE Enhanced (bonds, uncapped)trace_enhancedtrace_enhanced, trace_btds144a_enhanced, trace_agency_enhanced
TRACE Standard (bonds, capped)trace_standardtrace, trade_summary
FISD bond characteristics (Mergent)fisdfisd_mergedissue, fisd_mergedissuer, fisd_issue, fisd_issuer
TAQ Daily MSEC (ticks, µs)taqm_YYYYctm_YYYYMMDD, cqm_YYYYMMDD, complete_nbbo_YYYYMMDD, nbbom_YYYYMMDD, mastm_YYYY, wrds_iid_YYYY
TAQ Monthly (legacy, 1993–2014, sec)taqmonthly consolidated trades/quotes
CRSP↔TAQ linkwrdsapps_link_crsp_taqmlink table

Connection

Initialize PostgreSQL connection to WRDS:

import psycopg2

conn = psycopg2.connect(
    host='wrds-pgdata.wharton.upenn.edu',
    port=9737,
    database='wrds',
    sslmode='require'
    # Credentials from ~/.pgpass
)

Configure authentication via ~/.pgpass with chmod 600:

wrds-pgdata.wharton.upenn.edu:9737:wrds:USERNAME:PASSWORD

Connect via SSH tunnel:

ssh wrds

This uses ~/.ssh/wrds_rsa for authentication.

Critical Filters

Compustat Standard Filters

Always include for clean fundamental data:

WHERE indfmt = 'INDL'
  AND datafmt = 'STD'
  AND popsrc = 'D'
  AND consol = 'C'

CRSP v2 Common Stock Filter

Equivalent to legacy shrcd IN (10, 11):

df = df.loc[
    (df.sharetype == 'NS') &
    (df.securitytype == 'EQTY') &
    (df.securitysubtype == 'COM') &
    (df.usincflg == 'Y') &
    (df.issuertype.isin(['ACOR', 'CORP']))
]

Form 4 Transaction Types

WHERE acqdisp = 'D'  -- Dispositions
  AND trancode IN ('S', 'D', 'G', 'F')  -- Sales, Dispositions, Gifts, Tax

TAQ Standard Cleaning (Holden-Jacobsen 2014)

Minimum filter for academic intraday work:

-- Trades (ctm_YYYYMMDD)
WHERE tr_corr = '00'                               -- uncorrected
  AND price > 0 AND size > 0
  AND time_m BETWEEN '09:30:00'::time AND '16:00:00'::time
  AND (tr_scond IS NULL OR tr_scond = ''
       OR tr_scond !~ '[BGJKLOTUWZ4567]')          -- regular-way only

-- Quotes (cqm_YYYYMMDD)
WHERE bid > 0 AND ask > 0 AND ask > bid
  AND bidsiz > 0 AND asksiz > 0
  AND qu_cond IN ('A','B','H','O','R','W')
  AND time_m BETWEEN '09:30:00'::time AND '16:00:00'::time

Always prefer complete_nbbo_* over nbbom_* — the raw NBBO feed is incomplete when one exchange holds both sides of the top of book. If re-deriving, UNION nbbom with cqm records where natbbo_ind='1' OR nasdbbo_ind='4'. See references/taq.md.

TRACE Enhanced Cleaning (Dick-Nielsen 2014)

Apply BEFORE any price/volume analysis:

-- Post-2012-02-06: drop cancels, corrections, reversals + their matched originals.
-- Match originals on (cusip_id, entrd_vol_qt, rptd_pr, trd_exctn_dt,
-- trd_exctn_tm, rpt_side_cd, cntra_mp_id, msg_seq_nb).
WHERE trc_st = 'T'                              -- keep only clean originals
  AND NOT (cmsn_trd = 'Y' AND cntra_mp_id = 'D') -- drop agency double count
  AND spcl_trd_fl != 'Y'                         -- drop special-condition trades
  AND wis_fl != 'Y'                              -- drop when-issued
  -- For price/yield work, also:
  AND cmsn_trd = 'N'

Pre-2012-02-06 data uses asof_cd = 'R' for reversals instead of trc_st = 'Y'; use the pre-2012 branch. See references/trace.md.

Parameterized Queries

Always use parameterized queries (never string formatting):

Use scalar parameter binding for single values:

cursor.execute("""
    SELECT gvkey, conm FROM comp.company WHERE gvkey = %s
""", (gvkey,))

Use ANY() for list parameters:

cursor.execute("""
    SELECT * FROM comp.funda WHERE gvkey = ANY(%s)
""", (gvkey_list,))

Additional Resources

Reference Files

Detailed query patterns and table documentation:

  • references/compustat.md - Compustat tables, ExecuComp, financial variables
  • references/crsp.md - CRSP stock data, CCM linking, v2 format
  • references/insider-form4.md - Thomson Reuters Form 4, rolecodes, insider types
  • references/iss-compensation.md - ISS Incentive Lab, peer companies, compensation
  • references/edgar.md - SEC EDGAR filings, URL construction, DCN vs accession numbers
  • references/trace.md - TRACE bond transactions, Dick-Nielsen cleaning, FISD merge, pre-/post-2012 logic
  • references/taq.md - TAQ/TAQMSEC intraday trades, quotes, complete NBBO, Holden-Jacobsen filters, WRDS Intraday Indicators
  • references/connection.md - Connection pooling, caching, error handling

Example Files

Working code from real projects:

  • examples/form4_disposals.py - Insider trading analysis (from SVB project)
  • examples/wrds_connector.py - Connection pooling pattern

Scripts

  • scripts/test_connection.py - Validate WRDS connectivity

Local Sample Notebooks

WRDS-provided samples at ~/resources/wrds-code-samples/:

  • ResearchApps/CCM2025.ipynb - Modern CRSP-Compustat merge
  • ResearchApps/ff3_crspCIZ.ipynb - Fama-French factor construction
  • comp/sas/execcomp_ceo_screen.sas - ExecuComp patterns

Date Awareness

When querying historical data, leverage current date context for dynamic range calculations.

Current date is automatically available via datetime.now(). Apply this to:

  • Data range validation (e.g., "get data for last 5 years")
  • Fiscal year calculations
  • Event study windows

Implement dynamic date ranges in queries:

from datetime import datetime, timedelta

# Query last 5 years of data
end_date = datetime.now()
start_date = end_date - timedelta(days=5*365)

query = """
SELECT * FROM comp.funda
WHERE datadate BETWEEN %s AND %s
"""
df = pd.read_sql(query, conn, params=(start_date, end_date))

Always incorporate current date awareness in date-dependent queries to ensure results remain fresh across time.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.