Agent skill

Scholar RAG

by joshzyj in joshzyj/open-scholar-skill

Build and query a local vector database + GraphRAG over your entire reference library (Zotero or a PDF folder) for literature review.

Custom licenceAuto-check: notesResearch & Science

Install Scholar RAG

skills CLI
$ npx skills add joshzyj/open-scholar-skill --skill scholar-rag -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install joshzyj/open-scholar-skill scholar-rag --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/joshzyj/open-scholar-skill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/scholar-rag .claude/skills/scholar-rag && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
scholar-rag
GitHub stars
168
Token cost
~7.4k tokens
SKILL.md length
3,060 words
Files
23 (incl. references, assets)
Skills in repo
30
Repo updated
First seen
Licence
Custom licence

At a glance

Build and query a local vector database + GraphRAG over your entire reference library (Zotero or a PDF folder) for literature review.

  • Works in 4 steps: head — abstract + introduction → theory / literature / related work → discussion + conclusion → …
  • Build a searchable literature corpus
  • SKILL.md covers Arguments and Mode Routing, Setup block (run once per Bash…, MODE 0 — SETUP (one-time) and MODE 1 — INGEST (build the…, plus 11 more sections
  • Runs Python and Shell scripts from its folder; calls bash, python3 and brew

What it does

Scholar RAG is an agent skill from joshzyj/open-scholar-skill. Build and query a local vector database + GraphRAG over your entire reference library (Zotero or a PDF folder) for literature review. Downloads/locates full-text PDFs, extracts text (pdftotext/PyMuPDF/vision-OCR), chunks and embeds with bge-m3 into LanceDB, and layers a local-LLM GraphRAG (entity/relation extraction + Leiden communities + community summaries) seeded from the scholar-knowledge graph. Adds a bibliographic layer from OpenAlex (direct citations + bibliographic coupling + co-citation + paper-level…

Its SKILL.md is about 7.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 23 other files, including reference files and assets (for example `assets/_lib.sh`, `assets/build-all.sh` and `assets/chunk_embed.py`).

It sits in Research & Science, covering Knowledge graphs, Citation management and Retrieval-augmented generation. It works with Model Context Protocol, Ollama and Zotero. The repository describes itself as: Open scholar skill, a claude code plugin, for academic research.

When your agent uses it

  • Build a searchable literature corpus
  • Ground lit-review claims in cited passages
  • Find related papers

Example prompts

  • “/scholar-rag”

Requirements

  • Python 3
  • A Bash shell

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. head — abstract + introduction
  2. theory / literature / related work
  3. discussion + conclusion
  4. results + methods

What it can do on your machine

Read from SKILL.md and the folder at commit 6e5ac8e. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships script files (Python and Shell, from the files we listed), which the agent can run.

    Shell commands in SKILL.md call:

    • bash
    • python3
    • brew
    • claude

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Scholar RAG loads about 7.4k tokens when it runs, and up to ~12k if it reads all its reference files. Until then it costs about 231 tokens; SKILL.md has 3,060 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~231
When it runs · the whole SKILL.md, loaded when a task matches
~7.4k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~12k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NoteMentions a .env fileSKILL.md:364
    ## Configuration (`.env` or environment)
  • NoteMentions a .env fileSKILL.md:368
    These are read from `$SCHOLAR_SKILL_DIR/.env` and `~/.claude/.env` by `_lib.sh`.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 3,060 words (~7,411 tokens).

“You turn a scholar's whole reference library into a searchable, cited knowledge base. This skill ships a real, self-contained execution engine in assets/ (a Python package + a venv provisioner + a resumable build wrapper + an MCP server) —…”

— opening of SKILL.md by joshzyj, Custom licence
name
scholar-rag
tools
Read, Bash, Write
argument-hint
[setup|ingest|query|mcp|graph|citations|keywords|semantic|status] [args], e.g. 'ingest' or 'query how does segregation affect mobility' or 'graph run'
user-invocable
true

Read the full SKILL.md on GitHub

Files

SKILL.md and 22 other files (references, assets) in .claude/skills/scholar-rag of joshzyj/open-scholar-skill.

  • SKILL.md
  • assets/_lib.sh
  • assets/build-all.sh
  • assets/chunk_embed.py
  • assets/citations.py
  • assets/extract.py
  • assets/fetch.py
  • assets/graphrag.py
  • assets/ingest.py
  • assets/keywords.py
  • assets/mcp-setup.sh
  • assets/mcp_server.py
  • assets/query.py
  • assets/requirements.txt
  • assets/run-graph.sh
  • assets/run-ingest.sh
  • assets/semantic.py
  • assets/setup-venv.sh
  • assets/store.py
  • assets/sync-engine.sh
  • … and 3 more

Open the folder on GitHubat commit 6e5ac8e

Compare with similar skills

Scholar RAG next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Scholar RAG compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Scholar RAG this skilljoshzyj/open-scholar-skill168—~7.4kAutomated safety check: NotesCustom licence
Literature Review Toolsbrycewang-stanford/Auto-Empirical-Research-Skills4.5k—~2.5kAutomated safety check: NotesCustom licence
Aminer MCP ResearchDrchronx/ai-agent-research-starter-kit137—~1.1kAutomated safety check: PassCustom licence
Annotate Paper54yyyu/zotero-mcp5.3k—~1.5kAutomated safety check: PassMIT
NSFC Literature Review WriterHuiyuLi-2000/Chinese-Grant-Writer-Skills4341 repos~1.4kAutomated safety check: NotesMIT
Arxiv MCP Serverblazickjp/arxiv-mcp-server3.2k—~353Automated safety check: PassApache-2.0

Similar skills

  • Literature Review Tools

    brycewang-stanford/Auto-Empirical-Research-Skills

    Recommend AND run open-source AI tools, agents, Claude Code / Codex skills, and MCP servers for any stage of a literature review — searching, reading, extracting, synthesizing, screening…

    4.5k GitHub stars~2.5k tokensUpdated 4 days ago
    Research & ScienceAuto-check: notes
  • Aminer MCP Research

    Drchronx/ai-agent-research-starter-kit

    Use AMiner MCP for academic knowledge graph tasks: scholar search, author profile lookup, institution or team analysis, paper and patent discovery, academic influence checks, and research trend…

    137 GitHub stars~1.1k tokensUpdated 4 mo ago
    Research & ScienceAuto-check passed
  • Annotate Paper

    54yyyu/zotero-mcp

    Read the open paper and write study annotations into its PDF with zotero-cli - a context box on the title, a four-part summary on the abstract, role-coded abstract highlights, one box per figure…

    5.3k GitHub stars~1.5k tokensUpdated yesterday
    Research & ScienceAuto-check passed
  • NSFC Literature Review Writer

    HuiyuLi-2000/Chinese-Grant-Writer-Skills

    Writes the research-status literature review and critique section of an NSFC grant proposal, backed by a bundled multi-source literature search.

    434 GitHub starsUsed in 1 repo~1.4k tokens
    Research & ScienceAuto-check: notes
  • Arxiv MCP Server

    blazickjp/arxiv-mcp-server

    A skill your agent uses when finding, comparing, reading, or monitoring arXiv papers, including requests for abstracts, citation graphs, original LaTeX, section-level technical details, or…

    3.2k GitHub stars~353 tokensUpdated yesterday
    Research & ScienceAuto-check passed
  • Social Science Paper Writing

    fakerqwq/social-science-paper-writing-skill

    Helps draft, diagnose, review and revise social science papers, from topic and research question to literature review, citation risks and pre-submission checks.

    378 GitHub stars~7k tokensUpdated 4 mo ago
    Research & ScienceAuto-check passed

More from joshzyj/open-scholar-skill

All 30 skills in this repo
  • Scholar Annotate

    joshzyj/open-scholar-skill

    Turn unstructured text into validated, structured variables at corpus scale with LLMs: codebook design, dev/gold-set construction, DSPy prompt optimization, a hard reliability gate (Cohen κ ≥ 0.70)…

    168 GitHub stars~4.6k tokensUpdated 21 days ago
    Auto-check passed
  • Scholar Auto Research

    joshzyj/open-scholar-skill

    Stable, deterministic social-science research-paper pipeline from idea or data to verified manuscript, citations, replication package, and final md/docx/tex/pdf outputs.

    168 GitHub stars~21k tokensUpdated 21 days ago
    Auto-check passed
  • Scholar Causal

    joshzyj/open-scholar-skill

    Comprehensive causal inference toolkit for social science research.

    168 GitHub stars~10k tokensUpdated 21 days ago
    Auto-check passed
  • Scholar Data

    joshzyj/open-scholar-skill

    Comprehensive open data directory (100+ datasets across 14 categories) with auto-fetch capability, plus data collection instrument design, variable dictionaries, data management, IRB materials, and…

    168 GitHub stars~23k tokensUpdated 21 days ago
    Auto-check: notes
  • Scholar Eda

    joshzyj/open-scholar-skill

    Conduct exploratory data analysis (EDA) before hypothesis testing.

    168 GitHub stars~12k tokensUpdated 21 days ago
    Auto-check passed
  • Scholar Idea

    joshzyj/open-scholar-skill

    Explore broad social science ideas and convert them into formal, researchable questions.

    168 GitHub stars~8.8k tokensUpdated 21 days ago
    Auto-check: notes

Questions about Scholar RAG

What does Scholar RAG do?

Build and query a local vector database + GraphRAG over your entire reference library (Zotero or a PDF folder) for literature review. Scholar RAG is an agent skill from joshzyj/open-scholar-skill. Build and query a local vector database + GraphRAG over your entire reference library (Zotero or a PDF folder) for literature review.

When should I use Scholar RAG?

Scholar RAG fits situations like: build a searchable literature corpus; ground lit-review claims in cited passages; find related papers.

How do I install Scholar RAG in Claude Code?

Run `npx skills add joshzyj/open-scholar-skill --skill scholar-rag -a claude-code`. Or copy the skill folder (.claude/skills/scholar-rag in joshzyj/open-scholar-skill) into .claude/skills/scholar-rag in your project. Claude Code loads it when a task matches its description.

How do I install Scholar RAG in Codex?

Run `npx skills add joshzyj/open-scholar-skill --skill scholar-rag -a codex`. Or copy the skill folder (.claude/skills/scholar-rag in joshzyj/open-scholar-skill) into .agents/skills/scholar-rag in your project. Codex loads it when a task matches its description.

Can I use Scholar RAG in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add joshzyj/open-scholar-skill --skill scholar-rag -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/scholar-rag, .gemini/skills/scholar-rag, .github/skills/scholar-rag and .opencode/skills/scholar-rag in your project.

What does Scholar RAG need to run?

Going by SKILL.md and its folder, Scholar RAG needs Python and a shell for the scripts in its folder and the command-line tools its instructions call (bash, python3, brew and claude). Our summary lists: Python 3; A Bash shell.

Does Scholar RAG access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Scholar RAG safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Scholar RAG use?

Scholar RAG has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does Scholar RAG use?

About 7.4k tokens (SKILL.md is roughly 30k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 4.3k tokens, read only when the agent opens those files.

What are the alternatives to Scholar RAG?

Skills that share tags, products or a category with Scholar RAG: Literature Review Tools (brycewang-stanford/Auto-Empirical-Research-Skills, 4.5k stars), Aminer MCP Research (Drchronx/ai-agent-research-starter-kit, 137 stars), Annotate Paper (54yyyu/zotero-mcp, 5.3k stars) and NSFC Literature Review Writer (HuiyuLi-2000/Chinese-Grant-Writer-Skills, 434 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Scholar RAG?

joshzyj (a GitHub user) maintains it in joshzyj/open-scholar-skill, which has 168 GitHub stars. The repository holds 30 skills in this directory. The repository was last updated on September 18, 2026.

Source: joshzyj/open-scholar-skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.