dongzhigang13305312738-art/paper-skills / citation-assistant-skill
citation-assistant-skill
VerifiedClaude Code skill for citation workflow via OpenAlex and CrossRef
Citation Assistant Skill Guide
Overview
Citation Assistant is a Claude Code skill that integrates OpenAlex and CrossRef APIs into the coding workflow for instant paper lookup, citation formatting, and reference management. Search for papers by title or keyword, get formatted BibTeX entries, find related works, and insert citations — all without leaving the terminal. Designed for researchers writing papers in LaTeX or Markdown.
Installation
# Add as Claude Code skill
# Copy SKILL.md to your Claude Code skills directory
# Or install via OpenClaw:
openclaw skills install citation-assistant
Core Features
Paper Search
import requests
OA_API = "https://api.openalex.org"
def search_papers(query, limit=5):
"""Search OpenAlex for papers."""
resp = requests.get(
f"{OA_API}/works",
params={
"search": query,
"per_page": limit,
},
headers={"User-Agent": "ResearchPlugins/1.0 (https://wentor.ai)"},
)
return resp.json().get("results", [])
papers = search_papers("attention mechanism transformer")
for p in papers:
authors = [a["author"]["display_name"] for a in p.get("authorships", [])[:3]]
print(f"[{p.get('publication_year')}] {p.get('title')}")
print(f" {', '.join(authors)} — Citations: {p.get('cited_by_count')}")
print(f" DOI: {p.get('doi', 'N/A')}")
BibTeX Generation
def get_bibtex(doi):
"""Get BibTeX for a paper via CrossRef DOI resolution."""
resp = requests.get(
f"https://api.crossref.org/works/{doi}",
headers={"User-Agent": "ResearchPlugins/1.0 (https://wentor.ai; mailto:dev@wentor.ai)"},
)
msg = resp.json().get("message", {})
# Generate citation key
authors = msg.get("author", [])
first_author = authors[0].get("family", "unknown").lower() if authors else "unknown"
year = str(msg.get("published", {}).get("date-parts", [[""]])[0][0])
key = f"{first_author}{year}"
# Build BibTeX
authors_str = " and ".join(f"{a.get('given', '')} {a.get('family', '')}".strip() for a in authors)
doi_str = msg.get("DOI", "")
title = msg.get("title", [""])[0] if isinstance(msg.get("title"), list) else msg.get("title", "")
journal = msg.get("container-title", [""])[0] if msg.get("container-title") else ""
bibtex = f"""@article{{{key},
title = {{{title}}},
author = {{{authors_str}}},
year = {{{year}}},
journal = {{{journal}}},
doi = {{{doi_str}}},
}}"""
return bibtex
# Example
bibtex = get_bibtex("10.18653/v1/N19-1423")
print(bibtex)
Citation Context
def get_citing_works(openalex_id, limit=10):
"""Get papers that cite this work via OpenAlex."""
resp = requests.get(
f"{OA_API}/works",
params={
"filter": f"cites:{openalex_id}",
"per_page": limit,
"sort": "cited_by_count:desc",
},
headers={"User-Agent": "ResearchPlugins/1.0 (https://wentor.ai)"},
)
results = resp.json().get("results", [])
for paper in results:
authors = [a["author"]["display_name"] for a in paper.get("authorships", [])[:3]]
print(f"\n{paper.get('title')} ({paper.get('publication_year', '?')})")
print(f" Authors: {', '.join(authors)}")
print(f" Citations: {paper.get('cited_by_count', 0)}")
get_citing_works("W2741809807")
Related Paper Discovery
def find_related(openalex_id, limit=10):
"""Find papers related to a given paper via OpenAlex."""
# Get the paper's concepts, then search for similar works
resp = requests.get(
f"{OA_API}/works/{openalex_id}",
headers={"User-Agent": "ResearchPlugins/1.0 (https://wentor.ai)"},
)
paper = resp.json()
concepts = [c["display_name"] for c in paper.get("concepts", [])[:3]]
related_resp = requests.get(
f"{OA_API}/works",
params={
"search": " ".join(concepts),
"per_page": limit,
"sort": "cited_by_count:desc",
},
headers={"User-Agent": "ResearchPlugins/1.0 (https://wentor.ai)"},
)
return related_resp.json().get("results", [])
related = find_related("W2741809807")
for p in related:
print(f"[{p.get('publication_year')}] {p.get('title')} ({p.get('cited_by_count')} cites)")
Workflow Integration
### LaTeX Workflow
1. Search: "Find papers on transformer efficiency"
2. Select relevant papers from results
3. Generate BibTeX entries → append to references.bib
4. Insert \cite{key} in your .tex file
### Markdown Workflow
1. Search for papers while writing
2. Get formatted citation (APA, MLA, etc.)
3. Insert inline: (Author, Year) or [1]
4. Generate reference list at document end
Batch Operations
def build_bibliography(queries, output_file="refs.bib"):
"""Build BibTeX file from multiple search queries."""
all_bibtex = []
seen_ids = set()
for query in queries:
papers = search_papers(query, limit=3)
for paper in papers:
doi = paper.get("doi")
if doi and doi not in seen_ids:
seen_ids.add(doi)
bibtex = get_bibtex(doi.replace("https://doi.org/", ""))
all_bibtex.append(bibtex)
with open(output_file, "w") as f:
f.write("\n\n".join(all_bibtex))
print(f"Wrote {len(all_bibtex)} entries to {output_file}")
build_bibliography([
"attention mechanism",
"transformer architecture",
"BERT pre-training",
])
Use Cases
- Paper writing: Look up and format citations inline
- Literature review: Discover related papers from seed papers
- Reference management: Build BibTeX files from searches
- Citation analysis: Explore how papers cite each other
- Quick lookup: Find DOI, venue, citation count for any paper
References
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |