Objective PubMed Search Builder

Codex skill name: objective-pubmed

This repository contains a Codex skill for building objectively derived PubMed Boolean search strategies. It adapts a Hausner-style quasi-gold-standard workflow: confirm known relevant PMIDs, split development and held-out validation sets, mine candidate title/abstract and MeSH terms, build PubMed concept blocks, validate the strategy, and preserve an audit trail.

Method is based on Hausner E, Waffenschmidt S, Kaiser T, Simon M. Routine development of objectively derived search strategies. Syst Rev. 2012 Feb 29;1:19. doi: 10.1186/2046-4053-1-19. PMID: 22587829; PMCID: PMC3351720. + some divergences.

Adapts the PubMed, MeSH tools from https://github.com/aarontaycheehsien/pubmed-search-builder

The skill produces a documented PubMed strategy. It does not answer the evidence question or treat unconfirmed PubMed hits, similar articles, citation links, or candidate reviews as included studies.

Install

Clone the repository into your Codex skills directory using the installed skill folder name:

git clone https://github.com/aarontaycheehsien/objective-pubmed-search-builder.git ~/.codex/skills/objective-pubmed

On Windows, clone to:

git clone https://github.com/aarontaycheehsien/objective-pubmed-search-builder.git $env:USERPROFILE\.codex\skills\objective-pubmed

Then invoke it as:

Use $objective-pubmed to build and validate an objectively derived PubMed Boolean search from included-study PMIDs.

Requirements

• Python 3.10 or newer
• No third-party Python packages are required for the bundled scripts or tests
• Network access for PubMed E-utilities and MeSH RDF commands

Optional NCBI environment variables:

NCBI_EMAIL=<email>
NCBI_API_KEY=<api key>
NCBI_TOOL=objective-pubmed

Validate

Run the offline test suite:

python -m unittest discover -s tests -p "test_*.py"
python scripts/pubmed_tool.py selftest

Method Basis

• Hausner E, Waffenschmidt S, Kaiser T, Simon M. Routine development of objectively derived search strategies. Systematic Reviews. 2012;1:19. PMID: 22587829.
• Rethlefsen ML, Kirtley S, Waffenschmidt S, Ayala AP, Moher D, Page MJ, Koffel JB; PRISMA-S Group. PRISMA-S: an extension to the PRISMA Statement for Reporting Literature Searches in Systematic Reviews. Systematic Reviews. 2021;10:39. PMID: 33499930.