OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
MCP & Tooling

Paper Search MCP

Paper Search MCP is a Python-based Model Context Protocol server and command-line tool for searching academic literature across multiple sources, downloading available papers and extracting text. It combines unified multi-source search and deduplication with source-specific connectors and fallback-based retrieval.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Paper Search MCP — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a research workflow needs multi-source academic paper discovery and retrieval through an MCP client, CLI, or the documented Claude Code skill.

Who it is for

  • Researchers who need to search across academic indexes and repositories.
  • Developers adding paper discovery and retrieval tools to an MCP-compatible assistant.
  • Claude Code users who want a CLI-backed paper-search skill.
  • Users building research workflows around open and public academic sources.

Documented uses

  • Search multiple academic sources for papers and deduplicate results.
  • Choose broad or faster CLI search presets depending on coverage and latency needs.
  • Download papers through available source-native and open-access links with fallbacks.
  • Extract text from downloaded papers for research workflows.
  • Extend source coverage through source-specific connectors.
Before you choose

Requirements & limits

The repository is public and MIT-licensed. The README specifies Python 3.10 or newer. API keys are optional overall, though individual sources may require or benefit from credentials; Unpaywall is disabled without its email setting, and the IEEE connector requires an API key to activate.

  • Broad CLI search covers more sources; fast and fastest presets reduce the source set when lower latency matters more than coverage.
  • Search and full-text retrieval are separate: finding metadata does not guarantee an available PDF.
  • The service prioritizes public sources and uses open-access discovery and DOI resolution in its download fallback chain.
  • Provider restrictions, rate limits, network failures or access checks can affect source results.
  • HTTP transports support OAuth protected-resource mode with an external authorization server; local HTTP without OAuth should not be exposed directly.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install the Claude Code skill

Set up the documented Paper Search CLI and Claude Code skill for paper-search requests.

Before you start

  • uv
  • Claude Code

Steps

  1. Install the CLI with the command below.

    Documented command
    uv tool install paper-search-mcp

  2. Create the skill directory with mkdir -p ~/.claude/skills/paper-search.

  3. Download the skill file from the project repository into ~/.claude/skills/paper-search/SKILL.md.

  4. Optionally create ~/.config/paper-search-mcp/.env for provider API keys.

Expected outcome

Claude Code can use the skill to invoke the paper-search CLI for documented search, download and read tasks.

Guide limitations

  • The guide documents installation and the skill's intended use, not a tested execution result.
  • Optional API keys and provider conditions vary; some sources may be affected by external access limits or instability.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
3 November 2026
Original website title
openags/paper-search-mcp · Official homepage title · retained in its source language
How we review sourcesSuggest a correction