Guide · Agent skill
How to use the Systematic Literature Review agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The systematic-review skill is a set of written instructions that tells an AI agent to run a literature review in a structured, repeatable way, following ideas from the PRISMA guidelines. It is useful when you are preparing a related-work section, a survey, or a prior-work check on a topic. This guide explains what the skill changes in the agent's behaviour, how to install it, how to ask for it in plain language, and what to check before you trust the results.
Systematic Literature Reviewaiming-lab/AutoResearchClaw · MIT Download the skill Quick start
- 1Download the ZIP. Download the systematic-review skill ZIP from this page.
- 2Unzip into a skills folder. Extract it into ~/.claude/skills/ for all projects, or into a project's .claude/skills/ folder for one project.
- 3Or upload in the Claude apps. In the Claude apps, open Settings, go to Capabilities, then Skills, and upload the skill.
- 4Make a literature request. Ask for a literature review or related-work survey on a named topic, and state its purpose.
- 5Set your criteria. Give or confirm inclusion and exclusion criteria, since the method requires them before searching.
- 6Review the screening. Check the criteria, queries and excluded papers the agent reports.
- 7Verify citations. Open each cited paper and confirm it exists and supports the stated findings.
What the systematic-review skill makes the agent do
A skill is a folder containing instructions that an agent reads when a task matches. This one covers the literature stage of research. When you ask for a literature review or related-work survey, the agent is told to follow a fixed method rather than summarising whatever it finds first.
The method has eight points. The agent defines inclusion and exclusion criteria before searching, searches several databases (the source names Semantic Scholar, arXiv and OpenAlex), and uses both broad and narrow queries. It screens by title and abstract before reading full text, extracts method, dataset, metrics and key findings, and synthesises gaps and opportunities instead of only listing summaries.
- Sets criteria before any search begins
- Searches multiple databases with broad and narrow queries
- Screens by title and abstract first, then full text
- Extracts method, dataset, metrics and key findings
- Prioritises recent, highly cited papers and at least one seminal paper per sub-topic
What the skill does not do
The skill does not fetch papers itself. It runs no scripts and uses no network access, according to its listing. It only shapes how the agent approaches the work. Whether the agent can actually query databases depends on the tools available in your environment, which the skill does not provide.
It is also not a formal PRISMA compliance tool. It borrows the PRISMA idea of transparent, criteria-driven screening. Clinical or medical reviews that need full PRISMA reporting, protocol registration and dual human screening need more than this skill.
Install the skill from the ZIP download
Download the ZIP of the systematic-review skill from this page. Unzip it into your user skills folder, ~/.claude/skills/, so it is available in every project. To limit it to one project, unzip it into that project's .claude/skills/ folder instead.
If you use the Claude apps rather than a local setup, you can upload the skill instead. Open Settings, go to Capabilities, then Skills, and upload the skill there.
- Personal, all projects: ~/.claude/skills/
- Single project: the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload
Trigger the skill with natural requests
You do not need a special command. The skill's description and trigger keywords are literature, review, survey, related work and prior work. A request that uses these ideas should lead the agent to apply the method.
Be specific about the topic and the purpose. State what the review is for, such as a related-work section, so the agent knows what to extract and which gaps matter.
- Write a literature review on sparse attention methods for long-context language models.
- Survey prior work on evaluating LLM agents, with inclusion criteria I can check.
- Find related work for my paper on graph neural networks for molecular property prediction, and screen by abstract first.
- What are the open gaps in prior work on federated learning privacy?
Check and use the output
Treat the output as a draft for human checking. The agent may cite papers that do not exist or misstate findings. Open each cited paper, confirm it exists and says what the summary claims, and check the stated inclusion criteria against your own needs.
Ask the agent to show its criteria, search queries and the list of screened and excluded papers. This makes the process transparent and lets you repeat or audit it.
- Verify every citation against the original paper
- Confirm the criteria match your research question
- Ask for the queries and exclusion reasons
What people use it for
- Drafting the related-work section of a research paper on a specific method
- Building a survey of prior work before starting an experiment
- Screening a set of candidate papers by title and abstract with recorded exclusion reasons
- Identifying gaps across methods, datasets and metrics in a sub-field
- Running the literature stage of an AI research pipeline in a repeatable way
Frequently asked questions
Is the systematic-review skill free to use?
The skill is published under the MIT licence, as stated on its listing. Check the licence file in the source repository for the full terms before redistributing it.
What do I need to run the systematic-review skill?
The listing states no required dependencies. The skill runs no scripts and uses no network access, but the agent needs its own way to search databases for real results.
How do I install the systematic-review skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for all projects, or into a project's .claude/skills/ folder. In the Claude apps, you can instead upload it under Settings, Capabilities, Skills.
Is this skill a replacement for a PRISMA systematic review tool?
No. It borrows the PRISMA idea of criteria-driven, transparent screening but is not a complete compliance tool. Medical or clinical reviews needing formal PRISMA reporting, protocol registration and dual human screening need more than this skill.
Can the agent's literature results be trusted without checking?
No. The agent may cite papers that do not exist or misstate findings, so a human should verify each cited paper before using the review.
How does this compare with just asking an AI to summarise papers?
The skill forces a fixed process: criteria before searching, multiple databases, title and abstract screening before full text, structured extraction and gap synthesis. A plain summary request usually skips these steps.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
This skill suits researchers and students who want an AI agent to follow a disciplined, transparent process for a related-work survey, and teams with a research pipeline that needs a structured literature stage. It is not suitable for formal clinical systematic reviews, nor for use without a human checking the agent's citations and findings.
Download the skill