literature-review-writing
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-review-writing":https://skillpi.com/zh/skills/deepskill-e5bf52f8aae425cb27f06b4b1ef80e4bf5e86563045bd1662bddfeac08ac6cb3。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/writing/composition/literature-review-writing/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/writing/composition/literature-review-writing --skill literature-review-writing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-review-writing”技能的目录入口。公开标签:learning、文献综述、学术写作、系统性综述。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-review-writing
description: "Structure and write comprehensive literature reviews for any field"
metadata:
openclaw:
emoji: "📜"
category: "writing"
subcategory: "composition"
keywords: ["literature review", "systematic review", "narrative review", "synthesis", "academic writing", "review paper"]
source: "wentor-research-plugins"
---
# Literature Review Writing
A skill for structuring, synthesizing, and writing literature reviews. Covers narrative reviews, systematic reviews, scoping reviews, and literature review sections within empirical papers. Provides frameworks for organizing sources, identifying themes, and writing with synthesis rather than summary.
## Types of Literature Reviews
### Choosing the Right Review Type
| Type | Purpose | Search | Analysis | Length |
|------|---------|--------|----------|--------|
| Narrative | Broad overview of a topic | Selective | Qualitative synthesis | 5-30 pages |
| Systematic | Answer a specific question exhaustively | Exhaustive, documented | May include meta-analysis | 10-40 pages |
| Scoping | Map the extent of research on a topic | Broad, systematic | Charting and categorizing | 10-20 pages |
| Integrative | Synthesize diverse methodologies | Targeted | Conceptual synthesis | 10-25 pages |
| Umbrella | Review of systematic reviews | Focused on SRs | Synthesis of syntheses | 10-20 pages |
## Organizing Your Sources
### Building a Literature Matrix
```python
def create_literature_matrix(sources: list[dict]) -> dict:
"""
Create a structured matrix for organizing reviewed literature.
Args:
sources: List of dicts with keys: author, year, title, method,
sample, key_findings, themes, limitations
"""
matrix = {
"headers": [
"Author (Year)", "Research Question", "Method",
"Sample/Data", "Key Findings", "Themes", "Limitations"
],
"rows": [],
"theme_index": {}
}
for src in sources:
row = {
"citation": f"{src['author']} ({src['year']})",
"question": src.get("research_question", ""),
"method": src.get("method", ""),
"sample": src.get("sample", ""),
"findings": src.get("key_findings", ""),
"themes": src.get("themes", []),
"limitations": src.get("limitations", "")
}
matrix["rows"].append(row)
for theme in src.get("themes", []):
matrix["theme_index"].setdefault(theme, []).append(
row["citation"]
)
return matrix
```
### Thematic vs. Chronological Organization
```
Chronological (rarely ideal for reviews):
"In 2010, Smith found X. Then in 2012, Jones found Y.
In 2015, Lee found Z."
Problem: Reads like an annotated bibliography, not a synthesis.
Thematic (recommended):
"Three factors have been identified as predictors of X.
First, [factor A] has been consistently supported (Smith, 2010;
Jones, 2012; Lee, 2015). Second, [factor B] shows mixed results..."
Advantage: Synthesizes findings around conceptual themes.
Methodological (useful for systematic reviews):
"Studies using qualitative methods (n=12) found [pattern],
while quantitative studies (n=25) reported [different pattern].
This methodological divide suggests..."
```
## Writing with Synthesis
### Summary vs. Synthesis
```
Summary (weak -- describes one paper at a time):
"Smith (2020) studied 200 undergraduates and found that sleep
quality predicted academic performance. Jones (2021) surveyed
150 graduate students and found a similar relationship."
Synthesis (strong -- integrates multiple sources around a point):
"Sleep quality has been consistently linked to academic performance
across both undergraduate (Smith, 2020; Lee, 2019) and graduate
(Jones, 2021) populations, with effect sizes ranging from r=0.25
to r=0.42. However, this relationship may be confounded by
socioeconomic factors (Park, 2022), which only two studies
controlled for."
```
### Synthesis Sentence Starters
```
Agreement:
"There is broad consensus that..."
"Multiple studies converge on the finding that..."
"This finding has been replicated across [contexts]..."
Disagreement:
"However, findings diverge regarding..."
"In contrast to the majority view, [author] argues..."
"The evidence is mixed, with some studies reporting [X] and others [Y]..."
Gap identification:
"Notably absent from this literature is..."
"While [aspect] has been well studied, [gap] remains unexplored..."
"No studies to date have examined [specific gap]..."
Transition:
"Taken together, these findings suggest..."
"Building on this body of work, recent studies have begun to..."
"This line of research has evolved from [earlier focus] to [current focus]..."
```
## Structuring the Review
### Standard Sections
```
1. Introduction (1-2 pages)
- Define the topic and scope
- Explain why this review is needed (gap, timeliness, controversy)
- State the review's objectives or research questions
2. Methods (for systematic/scoping reviews)
- Search strategy, databases, date range
- Inclusion/exclusion criteria
- Screening process (PRISMA flow diagram)
3. Body: Thematic Sections (bulk of the review)
- Each section covers a theme, construct, or sub-question
- Synthesize rather than summarize
- Use tables to compare studies when appropriate
4. Discussion / Synthesis
- What is the overall state of knowledge?
- Where do studies agree and disagree?
- What are the gaps?
5. Conclusion / Future Directions
- Summarize the key takeaways
- Propose a research agenda addressing identified gaps
```
## Common Pitfalls
### What Reviewers Look For
```
Problem: "Laundry list" structure
- Each paragraph describes one study in isolation
Fix: Group studies by theme and synthesize across them
Problem: Missing recent literature
- Review stops at 2020 in a fast-moving field
Fix: Search within the last 12 months before submission
Problem: Uncritical acceptance
- All studies treated as equally valid
Fix: Evaluate methodological quality and note study limitations
Problem: No conceptual framework
- Sources listed without a guiding structure
Fix: Start with a framework (theoretical, conceptual, or thematic map)
Problem: Omitting contradictory evidence
- Only citing studies that support the authors' position
Fix: Actively seek and discuss disconfirming evidence
```
## Tools for Literature Review Management
- **Zotero + ZotFile**: Organize PDFs, tag by theme, generate bibliographies
- **Covidence**: Systematic review screening and data extraction
- **Rayyan**: Free AI-assisted abstract screening for systematic reviews
- **Notion/Obsidian**: Build a literature matrix with linked notes
- **VOSviewer**: Bibliometric visualization of citation networks and keyword clusters
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
提供多种文献综述类型(叙述性、系统性、范围性等)的对比与选择指南
评分依据
frontmatter 中 description 明确且位置正确
评分依据
文档结构清晰,分章节涵盖综述类型、组织方法、写作技巧和工具推荐
评分依据
提供的 literature_matrix 函数逻辑完整,参数定义清晰,可直接用于文献组织
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。
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。