agentsclimarketplace

Multi source literature brief

Skill ma-compbio-lab/SkillFoundry/skills/scientific-knowledge/multi-source-literature-brief

A framework for discovering, compiling, and validating reusable skills for scientific agents.

Install
npx -y skills add ma-compbio-lab/SkillFoundry --skill multi-source-literature-brief

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Use this skill to query OpenAlex, Europe PMC, Crossref, and PubMed for the same topic and build a normalized literature brief. Do not use it as a full systematic-review workflow or citation-analysis engine.

SKILL.md

2.8 KB, 604 tokens by cl100k_base, as published. Nobody here has run it

Purpose

Build a compact cross-source literature brief for a query by aggregating multiple official scholarly APIs into one normalized JSON report.

When to use

  • You want a quick cross-check of literature coverage across general and biomedical indexes.
  • You need a reusable JSON brief before deeper paper triage or ranking.
  • You want a small overlap view by DOI or title across sources.

When not to use

  • You need exhaustive retrieval for a systematic review.
  • You need full-text mining, citation graphs, or advanced deduplication.
  • You need offline-only execution.

Inputs

  • Query string
  • Per-source result limit
  • Optional output path
  • Optional NCBI/OpenAlex etiquette fields such as --email or --mailto

Outputs

  • JSON report with per-source counts, normalized top results, and cross-source overlap candidates

Requirements

  • Python 3.10+
  • Network access to OpenAlex, Europe PMC, Crossref, and NCBI E-utilities

Procedure

  1. Run python3 skills/scientific-knowledge/multi-source-literature-brief/scripts/build_literature_brief.py --query "single-cell RNA-seq" --limit 2 --out skills/scientific-knowledge/multi-source-literature-brief/assets/single_cell_literature_brief.json.
  2. Inspect the sources section to compare the top normalized results from each API.
  3. Review overlap_summary to see which DOI or title candidates appear in more than one source.

Validation

  • The command exits successfully.
  • All four source blocks are present.
  • Each source returns at least one normalized result for a known query.

Failure modes and fixes

  • NCBI rate limiting: rerun after a short pause or pass --email; the wrapper already retries transient 429 responses.
  • Sparse Crossref hits: keep the query phrase concise because Crossref is title-oriented here.
  • Query drift across sources: use quoted phrases or a more specific biomedical term.

Safety and limits

  • This skill aggregates public bibliographic metadata only.
  • It does not infer scientific truth or replace manual review.

Example

  • python3 skills/scientific-knowledge/multi-source-literature-brief/scripts/build_literature_brief.py --query "CRISPR screen melanoma" --limit 1

Provenance

Related skills

  • openalex-literature-search
  • europepmc-method-triage
  • crossref-metadata-search
  • ncbi-pubmed-search

What ships with it: 8 files

20.8 KB alongside SKILL.md, 2 of them executable

assets/

examples/

scripts/

tests/

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.