AlphaXiv MCP connector for AI agents

AlphaXiv MCP tools for AI agents

CALL ANY TOOL

4 tools covering get, answer, discover.

alphaxivmcp_list_library

List library

List the user's alphaXiv library: their folders (bookmark collections) with folder_id, name, type, parent_id, sharing_status, and paper_count. Set include_papers to also list papers per folder. Pass paper_ids_or_urls to check which folders already contain specific papers. The default folders 'default-to-read', 'default-reading', and 'default-completed' represent reading status; a paper may belong to any number of folders at once.

Parameters

Name

Type

Required

Description

include_papers

boolean

Optional

Also list the papers inside each folder (capped per folder).

paper_ids_or_urls

array

Optional

Report which folders already contain each of these papers.

alphaxivmcp_get_researcher

Get researcher

alphaxivmcp_list_followed_researchers

List followed researchers

alphaxivmcp_get_paper_content

Get paper content

alphaxivmcp_create_folder

Create folder

alphaxivmcp_get_researcher_papers

Get researcher papers

alphaxivmcp_rename_folder

Rename folder

alphaxivmcp_discover_papers

Discover papers

alphaxivmcp_find_researchers

Find researchers

The first tool for researcher affiliations, organization rosters, who left an organization, career moves, and what researchers are doing now. It composes name or subject relevance, current affiliation or role, citation range, position history, and verified coauthorship. Never use this tool merely to resolve a name for another call; pass the name directly to that profile or relationship call. Advising and lab-membership questions are answered here: coauthored_with returns the joint-work record behind each candidate, and require_position_overlap narrows it to people who were at the same place at the same time. Position constraints bind to one LinkedIn experience row. Returns [SLUG=...] handles.

Parameters

Name

Type

Required

Description

affiliation

string

Optional

Current organization ('DeepSeek', 'UC Berkeley'). Every indexed spelling of it is matched, so write the name plainly and never spell out variants yourself. For a historical organization use position.affiliation, or include_past_affiliations as shorthand.

include_past_affiliations

boolean

Optional

Shorthand for a past position at `affiliation`, excluding researchers whose current affiliation still matches. Use `position` for role or date constraints.

limit

integer

Optional

Results per page. Defaults to 12, or 50 for a relationship search, whose rows are one line each.

max_citations

integer

Optional

Citation ceiling, e.g. to find rising researchers who are not yet famous.

min_citations

integer

Optional

Citation floor, to keep a roster to established researchers.

page

integer

Optional

1-based page over the same ordering. Every result says whether more matched beyond it, so page on when the answer needs the tail rather than re-running a narrower search. A page that reports no total is not a count of the roster; only the last page of a search establishes one.

position

object

Optional

Composable LinkedIn position-history filter. Affiliation, role, status, and every date bound apply to the same experience row. For a recent departure use a past position with ended_after and end-date ordering; combine with top-level affiliation or role for transitions.

query

string

Optional

A person's name only when identifying that person is the final task, or a research subject ('mechanistic interpretability'). Never use this to resolve an anchor for another call; pass the name directly there. Omit for structured roster, relationship, or career queries.

relationship

object

Optional

Coauthorship evidence from linked public alphaXiv papers, one line per coauthor: shared papers, what share of that person's own output the collaboration is, how often they lead a paper the anchor closes, the years it ran, and their current position. require_position_overlap composes it with position history to add shared time.

role

string

Optional

Current role to filter by, token-matched ('professor' matches assistant and associate professors; 'phd student'). Only current roles are indexed. Omit or leave empty when not filtering by role.

sort

string

Optional

'relevance' ranks a query, or relationship strength for a relationship-only search, otherwise citations. 'recent_activity' ranks by latest linked paper. 'relationship_strength' requires relationship.

topic

string

Optional

Research subject to answer from authorship of papers on it, for 'who works on X' and 'who does prominent work on X'. One concept in the vocabulary its own papers use : name the architecture, benchmark, or subfield, and spell out an acronym alongside it. Never concatenate several related concepts to widen the net.

topic_authors

string

Optional

Which byline slot `topic` weights: 'lead' for the students and postdocs who drive the work, 'senior' for the PIs who close the author list, 'any' for both.

alphaxivmcp_follow_researcher

Follow researcher

alphaxivmcp_resolve_researchers

Resolve researchers

alphaxivmcp_unfollow_researcher

Unfollow researcher

alphaxivmcp_answer_pdf_queries

Answer pdf queries

alphaxivmcp_save_papers_to_folder

Save papers to folder

alphaxivmcp_move_papers_between_folders

Move papers between folders

alphaxivmcp_read_files_from_github_repository

Read files from github repository

alphaxivmcp_delete_folder

Delete folder

alphaxivmcp_remove_papers_from_folder

Remove papers from folder

Build your Agent

Same auth pattern across every framework.

from langchain_mcp_adapters.client import MultiServerMCPClient
from scalekit import ScalekitClient

client = ScalekitClient(env_url=ENV_URL, client_id=CLIENT_ID, client_secret=SECRET)
token = client.agent.get_token(user_id="user_id", connector="alphaxivmcp")

mcp = MultiServerMCPClient({
    "alphaxivmcp": {
        "url": "https://mcp.scalekit.com/alphaxivmcp",
        "headers": {"Authorization": "Bearer " + token}
    }
})
tools = await mcp.get_tools()

import OpenAI from "openai";
import { ScalekitClient } from "@scalekit-sdk/node";

const client = new ScalekitClient({ envUrl, clientId, clientSecret });
const token = await client.agent.getToken({ userId: "user_id", connector: "alphaxivmcp" });

const openai = new OpenAI();
// Connect to MCP at https://mcp.scalekit.com/alphaxivmcp
// Pass: Authorization: Bearer + token

import Anthropic from "@anthropic-ai/sdk";
import { ScalekitClient } from "@scalekit-sdk/node";

const client = new ScalekitClient({ envUrl, clientId, clientSecret });
const token = await client.agent.getToken({ userId: "user_id", connector: "alphaxivmcp" });

const anthropic = new Anthropic();
// Connect to MCP at https://mcp.scalekit.com/alphaxivmcp
// Pass: Authorization: Bearer + token

from google.adk.agents import LlmAgent
from scalekit import ScalekitClient

client = ScalekitClient(env_url=ENV_URL, client_id=CLIENT_ID, client_secret=SECRET)
token = client.agent.get_token(user_id="user_id", connector="alphaxivmcp")
# Connect to MCP at https://mcp.scalekit.com/alphaxivmcp
# Pass: Authorization: Bearer + token

Try these prompts

Paste any prompt into your agent to get started.

Get started

Copy the prompt

Copied

Reads the contents of a file or directory from the paper’s codebase repository?

Copy the prompt

Copied

Get the content of an arXiv/alphaXiv paper as text?

Advanced

Copy the prompt

Copied

Discovers and ranks multiple candidate papers for a research topic?

Copy the prompt

Copied

Returns raw filtered page content from one PDF as XML?

SEE HOW AUTH WORKS

User authorises once. Every agent call after uses their token with scope enforcement.

Authorize

Your user connects

AlphaXiv MCP

once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access

Who:

user ‘A’

when:

Once per user

access:

Limited to user

Store

Their

AlphaXiv MCP

token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection

vault:

encrypted

scope:

per-user

tokens:

auto-refreshed

Resolve

When your agent calls a

AlphaXiv MCP

tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs

speed:

~40ms

check:

before every call

seen by:

nobody

Audit

Every

AlphaXiv MCP

tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it

history:

90 days

export:

SIEM-ready

logged:

every call

Test other agents

See the same per-user auth pattern across other connectors.

GTM and RevOps Teams

Competitive intelligence briefing agent

Scans Gong calls for competitor mentions, matches each one to its Notion battlecard, and DMs every affected rep a single Slack digest per cycle. Every call runs as the PMM who owns the briefing, never a shared bot.

GTM and RevOps Teams

Outbound prospecting agent

Searches Apollo for prospects matching your ICP, scores and ranks them, drafts personalized outreach in Gmail, and logs every send to Google Sheets. Mail goes out as the rep, not from a shared inbox.

GTM and RevOps Teams

Deal intelligence agent

Pulls recent Gong calls, scores deal risk with an LLM, cross-references the record in Attio, and DMs each owner their at-risk deals in Slack. Every read is scoped to that rep's own access.

Support and Ops Teams

Support triage agent

Fetches new Zendesk tickets, classifies them by type and urgency, searches the Notion knowledge base for an answer, and routes what it cannot resolve to Slack. Every call runs on the support agent's own delegated OAuth.

Why Scalekit

Secure your agent's access. Connectors ship in minutes

01.

Shared tokens break per-user analytics

A shared token looks fine in a demo. In production every call looks like a service account. Scalekit resolves the real user credential.

// shared token
 audit → bot_service_account

 // scalekit
 audit → user_abc ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

One connector today. Ten tomorrow.

“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”

Venu Madhav Kattagoni

Head of Engineering / Von

FAQs

Frequently Asked Questions

Does the agent access AlphaXiv as the user or as a shared key?
As the user. Each workspace member authorizes once and Scalekit resolves their credential at request time. Audit logs attribute every action to that user, not a shared service account.

Where is the AlphaXiv OAuth token stored?
In Scalekit's managed AES-256 token vault, namespaced per tenant. Refresh is automatic. Revocation is a single dashboard action. Tokens never appear in prompts, logs, or LLM context.

Can I limit what the agent is allowed to do in AlphaXiv?
Yes. Pass a tool name filter to listScopedTools so the AI agent only sees the subset you authorize. Pre-API-call scope checks block out-of-policy actions before the request reaches AlphaXiv.

What happens when a user revokes AlphaXiv access?
The connection is invalidated on the next tool call. Subsequent requests for that user fail closed with a clear error. Other users in the tenant remain unaffected. The event is logged for audit.

Are paper queries and reading history per user?
Yes. Paper discovery, PDF queries, and repository reads run with the authorizing user's AlphaXiv credential, so research trails stay personal to each scientist.

On this page

Tool and actions Framework Compatibility Starter Prompt Auth Lifecycle Try Other Agents Why Scalekit