arXiv Search

Overview

Searches arXiv, the world's largest open library of scientific preprints. Great when you want the agent to pull recent papers on a topic — machine learning, physics, biology — and get back titles, authors, and links to the PDFs.

How it works

Issues a query to the public arXiv API — either free-text across the full corpus or a direct ID lookup (e.g. '2308.12345'). Returns paper titles, authors, abstracts, submission dates, and direct PDF links. No API key is needed; rate limits apply as set by arXiv.

Example

When a user asks:

Find recent papers on retrieval-augmented generation.

the agent calls the tool:

arxiv_search(query="retrieval-augmented generation", maxResults=10)

and gets back: a list of preprints with titles, authors, abstracts, and direct PDF links.

Use it in a workflow

Wire this tool into a SwarmAI crew. Use the YAML DSL for declarative workflows, or the Java builder API when you want full programmatic control.

YAML DSL

# literature-review.yaml
name: literature-review-crew
process: SEQUENTIAL

agents:
  - id: librarian
    role: Research Librarian
    goal: Find recent preprints relevant to the research question
    tools:
      - arxiv_search

tasks:
  - id: literature-review-task
    agent: librarian
    description: Search arXiv for 'retrieval-augmented generation' and return the top 10 papers with titles, authors, and PDF links.

Java

import ai.intelliswarm.swarmai.agent.Agent;
import ai.intelliswarm.swarmai.task.Task;
import ai.intelliswarm.swarmai.swarm.Swarm;
import ai.intelliswarm.swarmai.swarm.SwarmOutput;
import ai.intelliswarm.swarmai.process.ProcessType;
import ai.intelliswarm.swarmai.tool.research.ArxivTool;
import org.springframework.ai.chat.client.ChatClient;
import org.springframework.beans.factory.annotation.Autowired;

@Autowired ChatClient chatClient;
@Autowired ArxivTool arxivTool;

Agent librarian = Agent.builder()
    .role("Research Librarian")
    .goal("Find recent preprints relevant to the research question")
    .chatClient(chatClient)
    .tool(arxivTool)
    .build();

Task librarianTask = Task.builder()
    .description("Search arXiv for 'retrieval-augmented generation' and return the top 10 papers with titles, authors, and PDF links.")
    .agent(librarian)
    .build();

SwarmOutput result = Swarm.builder()
    .agent(librarian)
    .task(librarianTask)
    .process(ProcessType.SEQUENTIAL)
    .build()
    .kickoff();

What it's good for

Real scenarios where agents put this tool to work.

Source

Implementation lives at swarmai-tools/src/main/java/ai/intelliswarm/swarmai/tool/research/ArxivTool.java in the swarm-ai repository.

Open arxiv_search on GitHub →