Literature Review Tools plugin
Plugin and catalog of 70+ open-source literature review tools: six-database search, open-access full text and a tool launcher.
- 50 ★ on the repository
- Last commit
Extended access
Why this level
- The launcher installs third-party tools into virtual environments and stores model keys in a shared `.env` file.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/literature-review-toolsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The launcher installs third-party tools into virtual environments and stores model keys in a shared `.env` file.
This is third-party code. Review the repository files before installing.
What it does
The repository combines a curated list of open-source AI tools for literature review with an installable skill `literature-review-tools`. The skill does three things. It searches six sources at once (OpenAlex, Crossref, Semantic Scholar, PubMed, Europe PMC, arXiv) with no install or keys, finds legal full text by DOI and marks closed papers as closed. It recommends a tool per stage: search, screening, extraction, reading, synthesis, citation check, writing. And it runs the chosen tool through `scripts/litrun.py`. The launcher installs each tool in its own environment under `~/.lit-review-tools/`. Runnable tools include MinerU, marker and docling for PDFs, PaperQA2 for cited answers, ASReview for PRISMA screening, GPT Researcher and STORM, and the arXiv, paper-search and Zotero MCP servers. Ready pipelines such as `topic-to-review` chain search and a cited answer.
Who it is for. Researchers and graduate students building a literature review who want the agent to pick a tool, find papers and fetch open-access full texts.
Good fit when
- You need recent papers on a topic and have no API keys
- A DOI must become a legally accessible PDF or full text
- You are choosing a tool to turn PDFs into Markdown for a language model
- You are checking whether a reference in a draft exists
- A corpus of papers has to be collected and asked a question with source citations
Not a fit when
- You want to bypass paywalls: the skill marks closed papers as closed and does not try to open them
- You expect a fully automatic review without a researcher: tools help stages but a human verifies the result
- You need one API library only: the skill is broader and pulls several environments
Example request
Find recent papers on active learning for systematic review screening, check the DOIs and fetch open-access full texts.Limitations
The search scripts run on the Python standard library and need no keys, but Semantic Scholar without a key often hits its limit and each API has its own failure modes, described in `reference/apis/`. Runnable tools bring their own dependencies and models, and some (PaperQA2, GPT Researcher) need language model keys in the shared `~/.lit-review-tools/.env`. The 70-tool catalog is browse-only and only selected tools are runnable. The repository is CC0-1.0, while cataloged projects have their own licenses, so read them before commercial use. Catalog metadata refreshes weekly through GitHub.
How to disable. Disable the `lit-review-agent-tools` plugin through `/plugin` or delete the skill folder from `~/.claude/skills/`. Tool environments and the shared key file live in `~/.lit-review-tools/` and are removed separately.
FAQ
How do I install the plugin in Claude Code?
Run `/plugin marketplace add brycewang-stanford/lit-review-agent-tools`, then `/plugin install lit-review-agent-tools@lit-review-marketplace`. Another route is copying the `skills/literature-review-tools` folder into `~/.claude/skills/`.
Do I need keys to search papers?
No. `fetch_papers.py` and `resolve_oa.py` use only the standard library and open APIs, and a key is only needed for some runnable tools such as PaperQA2.
What does `litrun.py` do?
It installs the chosen tool in its own environment, runs it by id and reads keys from one shared file. Commands: `list`, `doctor`, `info`, `env`, `run`, `mcp`, `ui` and `workflow`.
Will the skill open paywalled papers?
No. Closed works get a closed status in `oa_report.json` and the skill has no paywall bypass.
Which ready pipelines exist?
Among them `pdf-to-markdown`, `pdf-corpus-qa`, `topic-to-pdfs`, `topic-to-review`, `topic-to-corpus` and `topic-to-fulltext`. Each can be previewed with `--dry-run`.
Discussion
No questions or reviews yet.
Related
PluginsA CLI, Python library and MCP server that scores how ready a site is to be cited by ChatGPT, Perplexity and AI Overviews
A GEO and SEO skill for the Chinese market: page scoring, rewrites for AI citation, llms.txt and competitor comparison
Open SEO Crawler: an SEO crawler with MCP for agents
Open SEO Crawler
A free local SEO crawler with a site audit and AI bot check: agents start crawls and read the issues
Eight read-only SEO MCP servers
seo-tools-mcp
SERP, Wordstat, GSC, GA4, Yandex Webmaster, Metrika and A-Parser: eight read-only MCP servers with strict JSON and multi-account support for Claude Code
Skill and CLI for SEO audits and website builds: a scroll story, GPT images, metadata checks and deployment to free hosting
A free SEO audit inside Codex CLI: 26 sub-skills and 19 expert roles check technical SEO, content, schema and AI search
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Literature Review Tools (this card) | no | 50 | ||
| Paper Search MCP | no | key is optional | 3,062 | an MCP server for paper search and download; this one is a skill with a tool catalog, multi-source search and pipelines |
| literature-survey-skill | no | 116 | literature survey through NotebookLM; this one picks and runs open-source tools and searches without keys |
README in short
The README in awesome-list format groups tools by task and stage and shows the maintenance status and license of each. A separate file covers the skill, the launcher and six verified recipes with real output.