agentleFS
AndreasADLAndreasADL/ai-papermill/.claude/skills/literature-scan/SKILL.md

literature-scan

Skill0 starsChanged 48 days agoMITView on GitHub

Scan academic literature for a research question using free scholarly APIs (OpenAlex, Semantic Scholar, Crossref). Use at the start of any study, or when the user asks for related work, prior findings, or references.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. jonfunk21/co-scientist★ 4same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository6 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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:

    Only cite papers found via the APIs.

    line 50
  2. Distinctive:

    If abstracts are insufficient for a key paper, note that only the abstract was reviewed.

    line 51
  3. Distinctive:

    If the literature is thin (<5 relevant papers), say so — that itself is a finding.

    line 52
  4. Very specific:

    Before finishing: check references.bib parses (count @ entries = papers.json entries) and spot-check 3 DOIs resolve (HTTP 200/302).

    line 55

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Literature Scan

Goal: a focused review of what is known about the research question, with verified references. Output goes to literature/.

Procedure

  1. Derive 3–6 search queries from the research question: synonyms, construct names, measurement instruments (e.g., "social media use depression", "screen time depressive symptoms adolescents").
  2. Query APIs (all keyless; always include a mailto/User-Agent where supported; sleep ~1s between calls):
import requests, time

# OpenAlex — best coverage + citation counts + open access links
r = requests.get("https://api.openalex.org/works",
    params={"search": QUERY, "per-page": 25, "sort": "cited_by_count:desc",
            "filter": "from_publication_date:2010-01-01",
            "mailto": "user@example.com"})
# fields: title, publication_year, cited_by_count, doi, abstract_inverted_index, authorships

# Semantic Scholar — good abstracts + TLDRs
r = requests.get("https://api.semanticscholar.org/graph/v1/paper/search",
    params={"query": QUERY, "limit": 20,
            "fields": "title,abstract,year,citationCount,externalIds,authors,venue"})

# Crossref — for verifying DOIs / metadata
r = requests.get("https://api.crossref.org/works", params={"query": QUERY, "rows": 10})

Note: OpenAlex returns abstracts as abstract_inverted_index — reconstruct by placing each word at its index positions.

  1. Select ~15–30 relevant papers. Prioritize: meta-analyses and reviews first, then highly cited empirical studies, then recent (last 5 years) studies. Drop papers whose abstracts show they are off-topic.
  2. Write outputs:
    • literature/papers.json — raw selected records (title, authors, year, venue, DOI, citations, abstract).
    • literature/references.bib — BibTeX for every selected paper. Fetch via content negotiation: requests.get(f"https://doi.org/{doi}", headers={"Accept": "application/x-bibtex"}). If a DOI fails to resolve, discard the reference — never hand-write BibTeX from memory.
    • literature/lit_review.md — structured review:
      • What is established (with effect size ranges if reported in abstracts)
      • Contested findings / heterogeneity
      • Common designs, measures, and typical confounders controlled for
      • Gaps the current study could address
      • Implications for our design (variables to include, expected effect direction/size)

Rules

  • Only cite papers found via the APIs. Every claim in lit_review.md needs an inline [Author, Year, DOI].
  • If abstracts are insufficient for a key paper, note that only the abstract was reviewed. Do not pretend to have read the full text.
  • If the literature is thin (<5 relevant papers), say so — that itself is a finding.

Verification

Before finishing: check references.bib parses (count @ entries = papers.json entries) and spot-check 3 DOIs resolve (HTTP 200/302).

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.