Skill Detail

Run source-audited academic literature searches with Nature Academic Search

Use Nature Academic Search to plan, deduplicate, verify, graph, and export academic literature searches across multiple scholarly sources.

Research & ScrapingMulti-Framework
Research & Scraping Multi-Framework Security Reviewed
⭐ 216 GitHub stars
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill run-source-audited-academic-literature-searches-with-nature-academic-search Copy
Uses the third-party skills CLI, not an ASE-owned installer. Check your agent’s compatibility. This copies instructions; complete the upstream tool setup below separately.
At a glance
Tools required
nature-academic-search CLI or MCP server, uv or compatible Python tool installer, PubMed contact email, optional Codex plugin, Claude Code plugin, or DeepSeek Harness bundle
Install & setup
For local CLI/MCP use, run uv tool install nature-academic-search, set PUBMED_EMAIL, then run nature-academic-search install –client both –email researcher@example.com and nature-academic-search preflight. Codex and Claude Code plugin installation paths are documented in the upstream README.
Author
wp-a
Publisher
Individual
Last updated
Sep 9, 2026
Quick brief

Use Nature Academic Search when an agent needs to turn a research question or seed paper into a reproducible literature-search package instead of a loose list of citations. The workflow defines the question and source policy, queries CrossRef, PubMed, arXiv, OpenAlex, Europe PMC, and optional scholarly sources, deduplicates by strong identifiers, verifies citation fields, expands bounded reference or cited-by graphs, and exports auditable JSON, RIS, BibTeX, NBIB, ENW, or graph artifacts. Invoke this instead of using a normal search box or asking a model for references when the user needs evidence that can be rechecked: queried sources, skipped sources, source records, conflicts, verification status, and result fingerprints. The scope boundary is reproducible academic discovery and citation verification for agents; it is not a generic web-search card, a paper summarizer, a citation manager, or just an MCP server listing.

How it works

What this skill actually does

Inputs and prerequisites: nature-academic-search CLI or MCP server, uv or compatible Python tool installer, PubMed contact email, optional Codex plugin, Claude Code plugin, or DeepSeek Harness bundle.

Setup notes: For local CLI/MCP use, run uv tool install nature-academic-search, set PUBMED_EMAIL, then run nature-academic-search install –client both –email researcher@example.com and nature-academic-search preflight. Codex and Claude Code plugin installation paths are documented in the upstream README.

Source and verification boundary: use https://www.wpironman.top/2026/07/academic-paper-search-reproducible-literature-search/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.

Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.