AGENT SKILL

research-knowledge-graph

Turn a bibliography or literature corpus into a knowledge graph of papers, authors, venues, topics and citation links, then analyze it (citation clusters, key papers, bridging work, research gaps) or export it to Neo4j, Gephi, NetworkX, Obsidian or Markdown-graph tools such as graphify. Use when mapping a research field, building a citation network, finding influential or bridging papers, or visualizing how a literature connects. Zero-dependency builder with optional OpenAlex enrichment.

License 许可
MIT
最近更新
2026-11-08
Stars
★ 0
Forks
0

01 前言 Front Matter

name
research-knowledge-graph
license
MIT
metadata
{ "tags": "knowledge graph, citation network, bibliometrics, OpenAlex, Neo4j, Gephi, NetworkX, Obsidian, graphify, science mapping", "author": "Kalaris Labs (Sayan Chowdhury)", "version": "1.0", "category": "knowledge-and-rag", "maintainer": "Kalaris Labs" }
description
Turn a bibliography or literature corpus into a knowledge graph of papers, authors, venues, topics and citation links, then analyze it (citation clusters, key papers, bridging work, research gaps) or export it to Neo4j, Gephi, NetworkX, Obsidian or Markdown-graph tools such as graphify. Use when mapping a research field, building a citation network, finding influential or bridging papers, or visualizing how a literature connects. Zero-dependency builder with optional OpenAlex enrichment.
compatibility
Python 3.9+ standard library; network access to api.openalex.org for --openalex. networkx optional for analysis.

02 资源文件 1

03 SKILL.md 正文

Research Knowledge Graphs

A literature review is a graph problem in disguise: which papers everyone builds on, which clusters talk past each other, which work bridges them, and where the gaps are. This skill builds that graph from a reference library and hands it to whichever graph tool the user prefers.

Build

# Offline: authors, venues and keywords from the .bib itself
python scripts/build_graph.py references.bib --out kg/

# Enriched: disambiguated authors (OpenAlex IDs, ORCID), venues, topics, and
# citation edges between papers in the library
python scripts/build_graph.py references.bib --out kg/ --openalex --mailto you@university.edu

CSL-JSON (.json, exported by Zotero or reference-manager-interop) also works as input.

Output Use it with
graph.json NetworkX: nx.node_link_graph(json.load(f), edges="links"), or any JS graph lib
graph.graphml Gephi, yEd, Cytoscape, igraph, nx.read_graphml
nodes.csv, edges.csv Neo4j (neo4j-admin database import full --nodes=nodes.csv --relationships=edges.csv) or Gephi spreadsheet import
vault/ Obsidian (graph view), and Markdown-graph tools such as graphify, which read [[wikilinks]] as edges

Schema. Node type ∈ {paper, author, venue, topic}. Edge type ∈ {authored_by, published_in, about, cites}. Paper nodes carry doi, year, openalex, cited_by_count when known.

Expand beyond the library (snowballing)

Citation edges only connect papers you already have. To map a field:

  1. Seed with 5-20 core papers.
  2. Backward: collect each seed's referenced_works from OpenAlex. Forward: query https://api.openalex.org/works?filter=cites:W123... for citing works.
  3. Keep candidates cited by ≥2 seeds (co-citation), export them to .bib, rebuild.
  4. Stop when a round adds few new highly connected papers. Report the rounds and thresholds (reproducibility).

Analyze (NetworkX)

import json, networkx as nx
g = nx.node_link_graph(json.load(open("kg/graph.json")), edges="links")
papers = [n for n, d in g.nodes(data=True) if d["type"] == "paper"]
cit = g.edge_subgraph([(u, v) for u, v, d in g.edges(data=True) if d["type"] == "cites"]).copy()

influential = sorted(cit.in_degree(), key=lambda x: -x[1])[:10]            # most cited within corpus
bridges = sorted(nx.betweenness_centrality(cit.to_undirected()).items(), key=lambda x: -x[1])[:10]
communities = nx.community.louvain_communities(cit.to_undirected(), seed=0)  # research clusters

Interpretation guidelines:

  • High in-degree within the corpus = foundational for this literature (not globally important).
  • High betweenness = bridges clusters, often the best papers to read for synthesis sections.
  • Communities = candidate subsections of a review. Name each by its top topics and most cited paper.
  • Sparse links between clusters that share topics = a potential research gap worth stating (and verifying).
  • Co-authorship components show research groups. Beware name-based author merging without --openalex.

Visualize

  • Gephi: open graph.graphml → ForceAtlas2 → size by in-degree → color by modularity class.
  • Obsidian: open kg/vault as a vault and use Graph view, filtering by folder (papers/authors/topics).
  • Small graphs in documents: convert the citation subgraph to Mermaid (flowchart LR, one edge per cites).

Integrity rules

  • Never invent edges or papers. Every node and edge must come from the library or the OpenAlex record.
  • Verify surprising structure, such as an apparent gap, by reading the papers before claiming it.

Caveats to state in any write-up

Coverage depends on OpenAlex/Crossref metadata; citation edges are only as complete as referenced_works. Keyword/topic nodes are coarse. Always report the data source, retrieval date and filtering thresholds.

Related skills

paper-corpus-rag (full-text QA over the same papers), literature-review, systematic-review-prisma, networkx, reference-manager-interop, citation-verification.

04 同仓库更多 Skills

Tashima-Tarsh
Tashima-Tarsh/Disha6.6

youtube-full

Use when the user needs YouTube transcripts, video search, channel browsing, playlist extraction, or content monitoring. Trigger phrases: 'get the transcript for', 'search YouTube for', 'what are the latest videos on', 'list this playlist', 'monitor this channel', or any request involving a YouTube URL, video ID, or @handle. Do NOT use for downloading video or audio files, YouTube engagement data (likes, comments), or private/age-restricted videos.

图像与多媒体
★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

x-twitter-growth

X/Twitter growth engine for building audience, crafting viral content, and analyzing engagement. Use when the user wants to grow on X/Twitter, write tweets or threads, analyze their X profile, research competitors on X, plan a posting strategy, or optimize engagement. Complements social-content (generic multi-platform) with X-specific depth: algorithm mechanics, thread engineering, reply strategy, profile optimization, and competitive intelligence via web search.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

write-a-skill

Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, build, or author a new skill.

Skill与智能体开发
★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

wrap-up

Close out a launched Claude Managed Agent — recap every primitive the founder now owns, regenerate the single-file overview page, and suggest the next 1-2 upgrades. Use when the user says "wrap up", "close this out", "what do I own now", "give me the summary", "recap the agent", or when the orchestrator routes phase=wrap-up. primitives_inventory.py tables everything owned (agent, environment, session, memory, outcome, deployment); overview_page.py regenerates a self-contained ./my-agent/agent-overview.html; upgrade_suggester.py ranks the next moves from recorded deferrals plus standing hardening steps. Companion to run-without-you; the last stop before phase=done.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

workflow-builder

Design and write deterministic multi-agent workflow scripts (.js files in .claude/workflows/) for Claude Code's Workflow tool. Use when a user wants to build, create, author, scaffold, or run a custom Claude Code workflow, orchestrate sub-agents (fan-out, pipeline, loop, judge-panel), or automate a repeatable multi-step task across fresh-context agents.

Skill与智能体开发
★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

weekly-review

Use when someone wants to run a weekly review, close open loops, audit stalled projects and commitments, get their system back to trusted, restart a lapsed review habit, or says "/cs:weekly-review". Walks David Allen's three-phase loop — GET CLEAR, GET CURRENT, GET CREATIVE — with deterministic scripts that inventory open loops, gate the checklist with named gaps, and score commitment health 0-100.

★ 0