literature-mapping-guide
安装方式
先用 Prompt 核对来源,也可切换到安装命令或下载可用的安装包。
用 AI 助手安装
复制提示词,让助手先核对固定版本和依赖。
查看 Prompt收起 Prompt
先检查当前 Agent 是否已安装 SkillPI Skill Finder。如果尚未安装,请按官方仓库的说明安装并启用:https://github.com/Findskills/skillpi。然后使用 Skill Finder 打开此 SkillPI 详情页并定位 "literature-mapping-guide":https://skillpi.com/zh/skills/deepskill-637f362689461011add9b7b3248456c47776ff43adfdf4be4bf4b02931485aca。安装前核对上游来源和版本:https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/blob/bfa0a6e5a843babbf45a1ef1663fd5dc4f07c0ba/skills/43-wentorai-research-plugins/skills/literature/discovery/literature-mapping-guide/SKILL.md。来源不一致时停止并说明差异。核对通过后,用 Skill Finder 安装目标 Skill,并报告安装路径和来源版本。
此命令从 GitHub 来源仓库获取指定技能。运行前请先查看源文件。
npx skills add https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/tree/bfa0a6e5a843babbf45a1ef1663fd5dc4f07c0ba/skills/43-wentorai-research-plugins/skills/literature/discovery/literature-mapping-guide --skill literature-mapping-guide来源与配置
查看 GitHub 源码brycewang-stanford/Auto-Empirical-Research-Skills固定提交 bfa0a6e5a843babbf45a1ef1663fd5dc4f07c0ba
配置方式待核实
请查看原始来源中的适用工具、依赖和权限要求。
安装实测待补充
本条目面向文献与学术工具场景,提供“literature-mapping-guide”技能的目录入口。公开标签:learning、文献图谱、引用网络、相似性分析。请通过固定上游版本核对实际功能、依赖与权限;此页为 Skillπ 独立编写的用途介绍和原站链接,不代表运行安全保证。原文与附件依各自许可开放。
文件浏览
---
name: literature-mapping-guide
description: "Visual literature mapping and connected papers exploration"
metadata:
openclaw:
emoji: "🗺️"
category: "literature"
subcategory: "discovery"
keywords: ["literature map", "reference graph", "citation network", "related papers"]
source: "wentor-research-plugins"
---
# Literature Mapping Guide
Build visual maps of scholarly literature to understand research landscapes, identify clusters of related work, and discover hidden connections between papers.
## What Is Literature Mapping?
Literature mapping transforms flat lists of papers into interactive visual networks where nodes represent papers and edges represent citation or similarity relationships. This approach helps researchers:
- See the overall structure of a research field at a glance
- Identify seminal papers (highly connected nodes)
- Discover clusters and subfields
- Find bridging papers that connect disparate areas
- Spot gaps where no papers exist
## Tools for Visual Literature Mapping
### Connected Papers
Connected Papers (connectedpapers.com) builds a similarity graph around a seed paper using co-citation and bibliographic coupling analysis.
| Feature | Details |
|---------|---------|
| Input | Paper title, DOI, or URL |
| Graph type | Similarity (not direct citation) |
| Node size | Citation count |
| Node color | Publication year (darker = older) |
| Max nodes | ~40 per graph |
| Cost | Free: 5 graphs/month; Premium: unlimited |
How to use:
1. Enter a seed paper URL or title
2. The tool builds a graph of the most similar papers (regardless of direct citation links)
3. Click any node to see its abstract and bibliographic details
4. Use "Prior works" to find foundational papers and "Derivative works" for recent developments
5. Export the graph or paper list as BibTeX
### Litmaps
Litmaps (litmaps.com) creates dynamic, multi-seed citation maps that update as new papers are published.
Workflow:
1. Add one or more seed papers via DOI, title, or OpenAlex ID
2. The tool builds a citation graph showing how the papers are connected
3. Add "discover" nodes to expand the map with algorithmically suggested papers
4. Create "collections" to organize maps by topic
5. Set up alerts for new papers that connect to your existing map
### VOSviewer
VOSviewer (vosviewer.com) is a free desktop tool for constructing and visualizing bibliometric networks at scale.
```
# VOSviewer supports several network types:
# - Co-authorship networks
# - Co-citation networks
# - Bibliographic coupling networks
# - Co-occurrence of keywords
# - Citation networks
# Input formats:
# - Web of Science export files
# - Scopus CSV exports
# - Dimensions export files
# - RIS files from reference managers
# - CrossRef API queries (built-in)
```
Steps for VOSviewer analysis:
1. Export search results from Web of Science or Scopus (include cited references)
2. Open VOSviewer and select "Create a map based on bibliographic data"
3. Choose analysis type (e.g., co-citation of cited references)
4. Set thresholds (e.g., minimum 5 citations for a reference to appear)
5. VOSviewer automatically clusters nodes and applies colors
6. Explore clusters to understand subfield structure
### CiteSpace
CiteSpace (citespace.podia.com) specializes in detecting research fronts and intellectual turning points.
Key features:
- Burst detection: identifies keywords or references with sudden spikes in frequency
- Timeline visualization: shows how clusters evolve over time
- Betweenness centrality: highlights bridging papers between clusters
- Requires Java; works with Web of Science data exports
## Building a Custom Literature Map with Python
```python
import networkx as nx
import requests
from collections import defaultdict
def build_citation_graph(seed_ids, depth=1, max_per_level=20):
"""Build a directed citation graph from seed papers."""
G = nx.DiGraph()
visited = set()
queue = [(sid, 0) for sid in seed_ids]
while queue:
paper_id, level = queue.pop(0)
if paper_id in visited or level > depth:
continue
visited.add(paper_id)
# Get paper metadata
meta_resp = requests.get(
f"https://api.semanticscholar.org/graph/v1/paper/{paper_id}",
params={"fields": "title,year,citationCount"}
)
if meta_resp.status_code != 200:
continue
meta = meta_resp.json()
G.add_node(paper_id, title=meta.get("title", ""),
year=meta.get("year"), citations=meta.get("citationCount", 0))
# Get references (backward)
refs_resp = requests.get(
f"https://api.semanticscholar.org/graph/v1/paper/{paper_id}/references",
params={"fields": "title,year,citationCount", "limit": max_per_level}
)
if refs_resp.status_code == 200:
for ref in refs_resp.json().get("data", []):
cited = ref["citedPaper"]
if cited.get("paperId"):
G.add_node(cited["paperId"], title=cited.get("title", ""),
year=cited.get("year"), citations=cited.get("citationCount", 0))
G.add_edge(paper_id, cited["paperId"], relation="cites")
if level < depth:
queue.append((cited["paperId"], level + 1))
return G
# Build graph from 2 seed papers
seeds = ["DOI:10.1038/s41586-021-03819-2", "ARXIV:2005.14165"]
graph = build_citation_graph(seeds, depth=1, max_per_level=15)
print(f"Graph: {graph.number_of_nodes()} nodes, {graph.number_of_edges()} edges")
# Find most central papers
centrality = nx.betweenness_centrality(graph)
top_central = sorted(centrality.items(), key=lambda x: x[1], reverse=True)[:10]
for node_id, score in top_central:
title = graph.nodes[node_id].get("title", "Unknown")
print(f" Centrality={score:.3f}: {title}")
```
## Interpreting Literature Maps
| Visual Feature | Interpretation |
|---------------|---------------|
| Large cluster | Established subfield with many related papers |
| Small isolated cluster | Emerging or niche research area |
| Bridge node between clusters | Interdisciplinary or foundational paper |
| Dense interconnections | Mature area with extensive cross-referencing |
| Sparse area between clusters | Potential research gap or opportunity |
| Temporal gradient (old to new) | Evolution of ideas over time |
## Best Practices
1. **Start with multiple seeds**: Using 3-5 seed papers from different angles gives better coverage than a single seed.
2. **Combine tools**: Use Connected Papers for quick exploration, VOSviewer for large-scale analysis, and custom scripts for specific analyses.
3. **Iterate**: Literature mapping is not a one-shot process. Refine your seeds and parameters based on initial results.
4. **Export and annotate**: Save your maps and annotate clusters with descriptive labels for use in literature review sections.
5. **Check boundaries**: If your map only shows papers from one group or country, broaden your seeds to capture different perspectives.
版本历史
-
v2026.07
GitHub ↗
原版更新说明
## Highlights - **First tagged release.** Every gate is green: `make check` (validate + 183 unit tests + eval-harness + numeric benchmarks), validate-catalog, quality-evals, and OpenSSF Scorecard on `main`. - **Rigor double loop at 13 method families.** Each family closes taxonomy tag → eval scenario (does the agent avoid the method's known traps?) → deterministic numeric benchmark (does it recover the true effect?). New this cycle: heterogeneous treatment effects (CATE) and quantile treatment effects (QTE). - **Paper-WorkFlow flagship at 29/29 executable rigor gates**, now merged to its `main` with a drift-gated README badge — the credibility artifact is checked by CI, so it can never silently go stale ([RIGOR.md](https://github.com/brycewang-stanford/Paper-WorkFlow/blob/main/RIGOR.md)). - **1,150 vendored skills / 69 collections**, all license-audited, security-scanned, tracked in a machine-readable catalog with CI-enforced freshness, and importable as a single whole-repo router skill. - **Hardened weekly upstream syncs**: synced trees must pass the full gate before merge; this cycle both sync PRs were verified, patched (StatsPAI's SkillOpt execution-gate card restored), and merged. - **First-party plugin marketplace** (Python / R / Stata empirical-analysis plugins + guide) shipped in the previous cycle, now under a tagged version. ## Known Follow-Ups - Ecosystem benchmark comparison (AERS vs econometrics agents on the shared benchmark suite) — [PLAN-2026-07 week 3](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/blob/main/docs/PLAN-2026-07.md). - Decomposition (Oaxaca–Blinder) and shift-share (Bartik) method families (weeks 3–4). - GitHub Pages catalog site, README demo screenshots, five-language README consistency script (week 4). ## Catalog - Top-level collections: **69** - `SKILL.md` files: **1150** - Cataloged agent tools: **335** ([`tools/CATALOG.md`](../tools/CATALOG.md)) ## Methodological rigor - Method families in the coverage map: **13** ([`RIGOR_COVERAGE.md`](RIGOR_COVERAGE.md)) - Eval scenarios: **26** (138 rubric items, 122 auto-checkable) ([`../eval-harness/`](../eval-harness/README.md)) - Benchmark tasks: **13** (70 gold checks, 58 required) ([`../benchmark/`](../benchmark/README.md)) ### Method families (skills tagged) - `iv`: 27 - `rdd`: 25 - `panel-fe`: 19 - `staggered-did`: 17 - `bayesian`: 13 - `did`: 10 - `event-study`: 10 - `matching`: 10 - `synthetic-control`: 10 - `cate`: 8 - `dml`: 8 - `survival`: 3 - `quantile`: 2 ### Eval scenarios by category - `causal-identification`: 14 - `citation-hygiene`: 1 - `reproducibility`: 2 - `research-integrity`: 1 - `runtime-safety`: 1 - `statistical-validity`: 2 - `writing-compliance`: 3 - `writing-style`: 2 ## License buckets - AGPL-3.0: 1 - Apache-2.0: 1 - CC-BY-NC-4.0 (non-commercial): 1 - CC-BY-SA-4.0: 1 - CC-BY-SA-4.0 (repository default): 4 - GPL-3.0: 1 - MIT: 35 - MIT Non-Commercial: 1 - Mixed (per-skill; see folder): 2 - UNKNOWN - check upstream: 22 ## Quality gates - `make check` — catalog/provenance/audit/eval freshness, repo hygiene, Paper-WorkFlow gates, stdlib unit tests, eval-harness lint + graded smoke, and the strict numeric benchmark (reference-gold drift fails the build). - External links and OpenSSF Scorecard run on their scheduled workflows.
当前收录版本
0.0.1
质量评测
满分 100。前四项来自 DeepSkill 收录的质量报告。
评分依据
提供可视化文献图谱构建与关联论文探索功能,解决研究人员学术文献发现、领域结构分析、研究空白识别等核心需求
评分依据
description清晰描述核心功能,frontmatter中keywords覆盖多种变体(literature map, reference graph, citation network, related papers),边界明确
评分依据
结构完整,分What Is、Tools、Custom Python、Interpreting、Best Practices等Phase,含端到端示例(Python代码),frontmatter完整
评分依据
工具来源明确(官网链接),Python代码参数合理,覆盖主要异常(status_code检查),无不可逆操作
计算方式
取前四项原始分值的算术平均,四舍五入取整。可通过收录报告核对各项评分与说明,运行记录需单独查证。
权限与使用前检查
根据当前主文档自动整理,命中内容可能是示例或防护说明;不是实际权限声明,也未覆盖附件、外部依赖或运行行为。
未从主文档识别到上述常见操作,不能据此判断无需权限或没有风险。
文档引用的域名或地址
这些地址可能用于文档链接或示例,不是已批准的联网清单。使用前核对实际目的地及发送的数据。
- api.semanticscholar.org
安全评级与访问范围
risk · 静态证据核验,不代表运行安全保证。
使用前请由宿主核对并授权实际文件读写、外部服务和联网目的地;本站的核验不会授予任何运行权限。
用户讨论
还没有已核实的点评。