Skip to content
Back to search

literature-review-skills

"Guide literature reviews from question refinement to evidence-grounded drafting, including journalism and communication, library and information science, and other disciplines. Use for 文献综述、新闻传播、情报学、综述写作、研究空白, or grill me / 追问 / 模拟答辩 about a review."

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/xingtaxueshu--literature-review-skills/badge.svg)](/listings/xingtaxueshu--literature-review-skills)

HTML

<a href="/listings/xingtaxueshu--literature-review-skills"><img src="/api/listings/xingtaxueshu--literature-review-skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

You do not need to copy commands for each individual skill. This repository includes a root `SKILL.md` as the entry point, plus the five working skills under `skills/`.

Method 1: One-Line Install (Recommended)

If your environment supports the Agent Skills installer, run:

npx skills add https://github.com/xingtaxueshu/literature-review-skills

Follow the installer's prompts to select your agent and skills, then check that the files are present. Discovery and copying behavior can vary by installer version; see the validation limits below.

Method 2: Ask Your Agent To Install It

Ask Claude Code, Codex, Cursor, or any compatible coding agent:

Please find and install this skill set: https://github.com/xingtaxueshu/literature-review-skills

Ask the agent to inspect its skills directory, download the repository, and verify that it can read the entry point. Completion depends on its file and network permissions.

Method 3: Manual Clone

For Claude Code users:

git clone https://github.com/xingtaxueshu/literature-review-skills.git ~/.claude/skills/literature-review-skills

For Codex users:

git clone https://github.com/xingtaxueshu/literature-review-skills.git ~/.agents/skills/literature-review-skills

After installation, ask your agent to use the skill set:

Use literature-review-skills to help me turn my broad topic into a literature review question, modules, method comparison, contradiction map, and research gap.

The root entry reads the relevant stage instructions. Manual cloning does not necessarily register all five stages as individually callable skills. To invoke one by name, select it in the installer or place its complete folder, including `references/`, in your agent's skills directory:

Use refine-literature-question on this topic: rural e-commerce participation.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 63 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: literature-review-skills
  3. description: "Guide literature reviews from question refinement to evidence-grounded drafting, including journalism and communication, library and information science, and other disciplines. Use for 文献综述、新闻传播、情报学、综述写作、研究空白, or grill me / 追问 / 模拟答辩 about a review."
  4. metadata:
  5. version: "0.3.0"
  6. ---
  7. # Literature Review Skills
  8. This is the entry point for the five literature-review skills in this repository.
  9. Use it when a user wants help with a literature review but has not yet chosen a specific skill. Route the user to the smallest skill that matches the current stage of work.
  10. ## Start Here
  11. Read [discipline routing](references/discipline-routing.md), then only the matching discipline profile. Read [review workflow](references/review-workflow.md) for search, extraction, and drafting. Read [research interview](references/research-interview.md) when the user requests grill me, pressure-testing, or has unresolved scope decisions.
  12. Use the user's language. Reuse information already supplied. Establish review type, question, discipline, scope, and available evidence. A scoped conceptual or historical question is valid; do not force causal hypotheses onto every review.
  13. For a complete review, follow the five stages below and then the drafting workflow. With no sources, provide a search plan and outline, not literature findings. With sources, draft thematic synthesis with claim-to-source links and explicit limitations.
  14. ## Loading Stage Instructions
  15. Resolve paths relative to this SKILL.md, not the working directory. Open the selected file before executing its workflow; naming a skill is not enough:
  16. | Stage | File to read |
  17. |---|---|
  18. | Question | [refine-literature-question](skills/refine-literature-question/SKILL.md) |
  19. | Structure | [decompose-review-modules](skills/decompose-review-modules/SKILL.md) |
  20. | Methods | [compare-research-methods](skills/compare-research-methods/SKILL.md) |
  21. | Synthesis | [map-literature-contradictions](skills/map-literature-contradictions/SKILL.md) |
  22. | Gap | [locate-review-research-gap](skills/locate-review-research-gap/SKILL.md) |
  23. If an installer omitted these files, report which are missing and request the complete repository or installation of the named stage skills. Do not claim to have loaded absent instructions. Individual stage skills carry their own references and remain usable independently.
  24. ## Skill Map
  25. 1. `refine-literature-question`
  26. - Use when the user has a broad topic and needs a sharp one-sentence review question.
  27. 2. `decompose-review-modules`
  28. - Use when the user already has a core question and needs 3-5 review modules.
  29. 3. `compare-research-methods`
  30. - Use when the user needs to justify method choices or compare method routes.
  31. 4. `map-literature-contradictions`
  32. - Use when the user has conflicting findings, notes, claims, or abstracts and needs a structured synthesis.
  33. 5. `locate-review-research-gap`
  34. - Use when the user needs to turn the review ending into a concrete research gap or future study direction.
  35. ## Source Integrity
  36. Before starting, check whether real references (full records, abstracts, or notes) are already available. Ask only if this is unresolved and the user has not requested no questions; otherwise record the evidence limit and proceed with a plan. Follow `references/source-integrity.md`: use only sources the user provided or that are verified, tag each with a status (用户提供 / 已核验 / 未核验), and write "未查到" instead of completing authors, year, journal, or pages. If the user has no references yet, say so in the output and give a method framework without literature claims. If that file is not available, still apply these rules.
  37. Read `references/output-contract.md`. Final deliverables end with `## 待核实清单` and `## 给作者的问题`; draft sections use `## 段落草稿(需作者改写并核对来源)` in the position specified by the stage or review workflow. Intermediate interview turns are exempt. In no-question mode, retain the author-question heading with "无(按要求不追加提问)".
  38. ## Recommended Sequence
  39. If the user is starting from scratch, use the skills in this order:
  40. ```text
  41. refine-literature-question -> decompose-review-modules -> compare-research-methods -> map-literature-contradictions -> locate-review-research-gap
  42. ```
  43. Do not treat this workflow as ghostwriting. The skills should help the researcher clarify questions, organize evidence, and test logic. They should not fabricate sources, invent findings, or replace the researcher's academic judgment. Generated paragraphs are drafts for the author to rewrite and check against sources.
README.md · README · 169 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Literature Review Skills
  2. A practical five-skill workflow for planning, structuring, and sharpening literature reviews. It helps you refine a question, break the review into modules, compare methods, handle contradictions, and locate a defensible research gap. It does not write the review for you; you still make the academic decisions, verify the literature, and write the final argument.
  3. [中文说明](README.zh-CN.md)
  4. ## Who This Is For
  5. - Graduate students preparing a thesis proposal, course paper, or dissertation literature review.
  6. - Researchers who need to turn scattered readings into a clear review structure.
  7. - Supervisors who want a repeatable workflow for guiding students through review design.
  8. - Claude Code and Codex users who want reusable skills for academic thinking tasks.
  9. ## The Five-Skill Workflow
  10. | Step | Skill | Use it to |
  11. |---|---|---|
  12. | 1 | `refine-literature-question` | Turn a broad topic into a focused review question that guides search and screening. |
  13. | 2 | `decompose-review-modules` | Break the question into 3-5 review modules with clear boundaries. |
  14. | 3 | `compare-research-methods` | Compare method routes and explain what each method can and cannot answer. |
  15. | 4 | `map-literature-contradictions` | Turn conflicting findings into structured synthesis instead of vague disagreement. |
  16. | 5 | `locate-review-research-gap` | Convert the synthesis into a concrete, defensible research gap or future direction. |
  17. ## Disciplines And Drafting
  18. v0.3 keeps the five stages and adds discipline guides loaded only when relevant.
  19. | Discipline | Focus |
  20. | --- | --- |
  21. | Journalism and communication | News production, journalism, platforms, audiences, and credibility; distinguish text-, person-, organization-, and platform-level evidence. |
  22. | Library and information science | Information behavior, retrieval, knowledge organization, bibliometrics, scholarly communication, and services; separate system metrics, task outcomes, and user experience. |
  23. | Other or interdisciplinary topics | Use the general workflow and state when no dedicated guide exists. Combine two guides when needed without forcing one field's assumptions onto another. |
  24. “News” means scholarly journalism/communication reviews, not news digests. Ambiguous uses of “intelligence” require clarification; a security-intelligence guide is not included.
  25. The workflow supports planning and drafting narrative, scoping, systematic, and conceptual/historical reviews: define the question and review type, document searches and screening, extract evidence, synthesize, then draft with traceable claims. It does not claim to complete a systematic review from a prompt alone.
  26. ```text
  27. Use literature-review-skills for a journalism and communication review on AI-generated
  28. news disclosure and audience credibility judgments. Grill me one question at a time
  29. to settle concepts, scope, and review type. I have no papers yet; do not invent findings.
  30. ```
  31. ```text
  32. Use literature-review-skills for an information-science review of AI-assisted scholarly
  33. search. I will provide records, abstracts, and reading notes. Build an evidence table,
  34. distinguish retrieval performance from task success and user experience, then propose
  35. an outline. Do not rank scores from different datasets and metrics directly.
  36. ```
  37. Ask for “grill me” to work through one decision at a time, or “proceed without questions” to get a plan with explicit assumptions. Once evidence is available, ask for a draft based on the evidence table. Missing sources remain gaps to verify, not invented answers.
  38. See [usage examples](examples/discipline-workflows.md). To add a field, use the [discipline template](references/disciplines/_template.md), register it in the [routing table](references/discipline-routing.md), and add behavioral cases.
  39. ## Install This Skill Set
  40. You do not need to copy commands for each individual skill. This repository includes a root `SKILL.md` as the entry point, plus the five working skills under `skills/`.
  41. ### Method 1: One-Line Install (Recommended)
  42. If your environment supports the Agent Skills installer, run:
  43. ```bash
  44. npx skills add https://github.com/xingtaxueshu/literature-review-skills
  45. ```
  46. Follow the installer's prompts to select your agent and skills, then check that the files are present. Discovery and copying behavior can vary by installer version; see the validation limits below.
  47. ### Method 2: Ask Your Agent To Install It
  48. Ask Claude Code, Codex, Cursor, or any compatible coding agent:
  49. ```text
  50. Please find and install this skill set: https://github.com/xingtaxueshu/literature-review-skills
  51. ```
  52. Ask the agent to inspect its skills directory, download the repository, and verify that it can read the entry point. Completion depends on its file and network permissions.
  53. ### Method 3: Manual Clone
  54. For Claude Code users:
  55. ```bash
  56. git clone https://github.com/xingtaxueshu/literature-review-skills.git ~/.claude/skills/literature-review-skills
  57. ```
  58. For Codex users:
  59. ```bash
  60. git clone https://github.com/xingtaxueshu/literature-review-skills.git ~/.agents/skills/literature-review-skills
  61. ```
  62. After installation, ask your agent to use the skill set:
  63. ```text
  64. Use literature-review-skills to help me turn my broad topic into a literature review question, modules, method comparison, contradiction map, and research gap.
  65. ```
  66. The root entry reads the relevant stage instructions. Manual cloning does not necessarily register all five stages as individually callable skills. To invoke one by name, select it in the installer or place its complete folder, including `references/`, in your agent's skills directory:
  67. ```text
  68. Use refine-literature-question on this topic: rural e-commerce participation.
  69. ```
  70. ## Suggested Workflow
  71. Sample prompt:
  72. ```text
  73. I am writing a literature review on how rural e-commerce affects household income and local development. Help me turn this broad topic into a review question, modules, method comparison, contradiction map, and research gap.
  74. ```
  75. Recommended sequence:
  76. 1. Use `refine-literature-question` to convert the topic into a focused question, such as how rural e-commerce changes household income through market access, labor allocation, or platform participation.
  77. 2. Use `decompose-review-modules` to create review sections, such as adoption conditions, income mechanisms, platform roles, regional differences, and policy environment.
  78. 3. Use `compare-research-methods` to understand what surveys, panel data, interviews, case studies, and mixed methods can and cannot answer.
  79. 4. Use `map-literature-contradictions` to compare studies that report positive, weak, uneven, or negative effects and identify why results differ.
  80. 5. Use `locate-review-research-gap` to state what the literature explains, where it gets stuck, and what a next study could examine.
  81. ## Responsible Use
  82. These skills are a thinking workflow, not a ghostwriter. Use them to clarify questions, organize evidence, test logic, and improve research design. Do not use them to fabricate sources, outsource academic judgment, or submit generated prose without checking it against real literature and your own argument.
  83. Since v0.2.0 the skills follow [`references/source-integrity.md`](references/source-integrity.md): only sources you provided or that have been verified are used, each is tagged (用户提供 / 已核验 / 未核验), and anything not found is written as "未查到" rather than filled in. Final deliverables end with a to-verify list and an author-question section, and generated paragraphs are labelled as drafts for you to rewrite and check ([output contract](references/output-contract.md)). These are process rules for the agent, not a guarantee; check every citation yourself.
  84. ## Limitations And Unverified Items
  85. This is an early version. Please read the following before relying on it.
  86. - **No quality evaluation.** The repository includes [behavioral cases](evals/scenarios.json) and reference-sync tests, but no completed model-behavior evaluation, baseline comparison, or real-manuscript quality assessment. Benefits described here are design intent, not measured results.
  87. - **Usage feedback has not been collected systematically.** Earlier external use mainly came through WeChat posts. The v0.3 discipline guides and interview workflow have not been validated on real manuscripts.
  88. - **The example has no real sources.** `examples/` is a method demonstration; its claims are illustrative, not drawn from specific papers, and must not be cited.
  89. - **Install methods are not individually tested.** We have not verified the `npx skills add` command, agent-assisted install, or manual clone in each environment (Claude Code, Codex, Cursor). The `agents/openai.yaml` files are also untested. So that a skill still works if an installer copies only one skill folder, each skill folder bundles its own copy of `references/`, and the core rules are also repeated in each skill's Guardrails.
  90. - **Source verification depends on the agent's tools.** Without network or search access, the agent cannot verify anything and will leave sources as 用户提供 or 未核验. Crossref and OpenAlex coverage of Chinese-language and book literature is incomplete.
  91. - **Chinese databases still require authorized manual access.** Bilingual search planning, search logs, CSSCI filtering scope, and record checks are included, not automated database login or guaranteed citation-style compliance.
  92. - **Discipline guides are not comprehensive methods textbooks.** Conceptual/historical questions and field-specific appraisal checks are included, but need adjustment to the research tradition and corpus.
  93. - **Outputs need spot-checking.** Models may skip sections, mislabel sources, or ignore interview constraints. Intermediate interview turns are exempt from the full delivery template; final deliverables retain the verification and author-question sections.
  94. ## Contact
  95. Questions, examples, or collaboration ideas are welcome. Please use [GitHub Issues](https://github.com/xingtaxueshu/literature-review-skills/issues/new/choose) for usage feedback.
  96. - WeChat: `xingtacc`
  97. - Email: [xingtaxueshu@outlook.com](mailto:xingtaxueshu@outlook.com)
  98. Scan the QR code to add the WeChat account:
  99. <img src="assets/wechat-qr.jpg" alt="WeChat QR code for 星塔学术" width="260">
  100. Follow the WeChat Official Account **星塔学术** (scan the QR code, or search for it in WeChat) for walkthroughs and updates:
  101. <img src="assets/wechat-official-account-qr.png" alt="WeChat Official Account QR code for 星塔学术 · StarTower Academic" width="260">
  102. ## Contributing
  103. Contributions are welcome if they make the workflow clearer, more practical, or easier to use. Good contributions include sharper skill instructions, better examples, field-specific templates, and corrections to installation notes. Keep examples realistic and avoid inflated claims.
  104. `skills/*/references/` are copies of the authoritative root `references/`. After editing, run `python3 scripts/sync_references.py`, then `bash scripts/check_references_sync.sh`. The check includes nested discipline files and unexpected copies.
  105. Design references include grill-me / grilling for interviewing, K-Dense literature-review for search and verification, and GPT Researcher for separating plans, source material, and reports. See [design references](docs/design-references.md) for scope and attribution. None is a runtime dependency.
  106. ## Related Projects
  107. - [xingta-polish-en](https://github.com/xingtaxueshu/xingta-polish-en): a fidelity-first agent skill for polishing the English of academic paragraphs, from the same team (星塔学术 · StarTower Academic).
  108. ## License
  109. This project is released under the MIT License. See `LICENSE` for the full text. Maintained by 星塔学术 · StarTower Academic ([www.xingta.cc](https://www.xingta.cc)).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.