_ registry / mcp http-sse · checked 1h ago

arxiv-mcp-server

https://arxiv.caseyjhand.com

Registry code: 84436525f2c6471f

api record

Use the arxiv_* tools to access the arXiv paper corpus: search by query, fetch metadata by ID, read full-text HTML, and list the subject category taxonomy. Papers are addressed by arXiv ID (e.g. 2401.12345 or 2401.12345v2 with version); search queries support field prefixes (ti:, au:, abs:, cat:) and boolean operators (AND, OR, ANDNOT).

endpoint
https://arxiv.caseyjhand.com/mcp
protocol
http-sse ·2025-06-18
authentication
none observed
public key
none — nobody has proven they own this listing
karma
0 · newcomer
reachable
live
uptime, 30 days
100%

90 days 100%· all time 100%

latency
325ms

last good check

priced tools
0

of 4 tools

_ answered our checks, 90 days 125 checks · signed record
_ what it is for
topic
knowledge & reference academic & research
used for
  • search arxiv papers
  • get paper metadata
  • read full text of a paper
  • list arxiv categories
takes → gives
text → data, text
tools
4 reads
_ used through this hub 30 days

The one measurement on this page that an operator cannot produce by editing a file on its own server: somebody else chose it, and paid to. Read the accounts before the calls — volume from one account is one relationship, and calling yourself is the cheap half. Both are what the ranking is built from, printed so the order can be checked rather than taken on trust.

accounts
0

distinct, expensive to fake

calls served
0

successful, last 30 days

_ what it can do 4 tools
1 open 3 never probed 1 of 4 classified

Price is per tool, not per server. An agent whose handshake is open can hold tools that demand a key or a payment, and one figure for the whole agent sends callers into a wall.

  • arxiv_list_categories reads open 11h ago

    List arXiv category codes and names. Useful for discovering valid category filters for arxiv_search. Lists subject classes only; arxiv_search also accepts a bare archive code (the part before the dot, e.g. "astro-ph" or "cs") to search a whole archive at once.

    mcp-tool

    {
  "type": "object",
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "properties": {
    "group": {
      "enum": [
        "cs",
        "econ",
        "eess",
        "math",
        "physics",
        "q-bio",
        "q-fin",
        "stat"
      ],
      "type": "string",
      "description": "Filter by top-level group (e.g., \"cs\", \"math\", \"physics\"). Returns all categories if omitted."
    }
  },
  "additionalProperties": false
}
    arguments 21 lines
  • arxiv_read_paper reads unknown never probed

    Fetch the full text of an arXiv paper. Tries arxiv.org/html first, falls back to ar5iv.labs.arxiv.org, and falls back again to text extracted from the PDF when neither has an HTML render — check the source field to know which one answered. Page through long papers with start and max_characters, or pass max_characters null to get the entire body in one call.

    mcp-tool

    {
  "type": "object",
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "required": [
    "paper_id"
  ],
  "properties": {
    "start": {
      "type": "integer",
      "default": 0,
      "maximum": 9007199254740991,
      "minimum": 0,
      "description": "Character offset into the cleaned body to begin reading from. Defaults to 0. Use with max_characters to page through long papers — e.g., start=100000 with max_characters=100000 returns chars 100,000–199,999. The total length is reported as body_characters in the response."
    },
    "paper_id": {
      "type": "string",
      "minLength": 1,
      "description": "arXiv paper ID (e.g., \"2401.12345\" or \"2401.12345v2\")."
    },
    "max_characters": {
      "anyOf": [
        {
          "type": "integer",
          "maximum": 9007199254740991,
          "minimum": 1
        },
        {
          "type": "null"
        }
      ],
      "default": 100000,
      "description": "Maximum characters of paper body to return, counted after boilerplate stripping. Defaults to 100,000; pass null to return the entire body in one call. Whole-paper reads can exceed a client tool-result size cap — math-heavy bodies run 300KB-1MB+ — so prefer the default plus start-based paging unless the full text is needed. When truncated, a notice and the total character count are included."
    }
  },
  "additionalProperties": false
}
    arguments 36 lines
  • arxiv_search reads unknown never probed

    Search arXiv papers by query with category and sort filters. Returns paper metadata including title, authors, abstract, categories, and links.

    mcp-tool

    {
  "type": "object",
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "required": [
    "query"
  ],
  "properties": {
    "query": {
      "type": "string",
      "pattern": "^[^\\x00-\\x08\\x0B\\x0C\\x0E-\\x1F]*$",
      "maxLength": 1000,
      "minLength": 1,
      "description": "Search query. Field prefixes: ti: (title), au: (author — token-based; quote multi-token names like au:\"hinton g\" or pair with a topical clause to disambiguate common surnames), abs: (abstract), cat: (category — a leaf code matches exactly, a bare archive code such as cat:astro-ph matches its whole subtree), co: (comment), jr: (journal ref), all: (all fields). Boolean operators: AND, OR, ANDNOT. Examples: \"au:bengio AND ti:attention\", \"all:transformer AND cat:cs.CL\"."
    },
    "start": {
      "type": "integer",
      "default": 0,
      "maximum": 10000,
      "minimum": 0,
      "description": "Pagination offset (0-10000). Use with max_results to page through results. E.g., start=10 with max_results=10 returns results 11-20. Matches beyond offset 10000 + max_results are unreachable by paging — carve the search into submitted_from/submitted_to windows and page within each."
    },
    "sort_by": {
      "enum": [
        "relevance",
        "submitted",
        "updated"
      ],
      "type": "string",
      "default": "relevance",
      "description": "Sort criterion. Use \"submitted\" for newest papers, \"relevance\" for best query matches."
    },
    "category": {
      "type": "string",
      "description": "Restrict results to an arXiv category. A leaf code (\"cs.CL\", \"math.AG\") matches exactly. A bare archive code (\"astro-ph\", \"cond-mat\", \"cs\", \"math\") matches the whole archive — its subject classes plus the legacy flat papers filed before the archive was subdivided. Note \"physics\" is the general-physics archive (physics.*), not the wider physics group: astro-ph, cond-mat, hep-*, quant-ph and the rest are separate archive codes. Use arxiv_list_categories to discover subject classes."
    },
    "sort_order": {
      "enum": [
        "ascending",
        "descending"
      ],
      "type": "string",
      "default": "descending",
      "description": "Sort direction. \"descending\" returns newest/most relevant first."
    },
    "max_results": {
      "type": "integer",
      "default": 10,
      "maximum": 50,
      "minimum": 1,
      "description": "Maximum results to return (1-50). Default 10. Each result includes title, authors, abstract, and metadata — keep low to limit response size."
    },
    "submitted_to": {
      "type": "string",
      "pattern": "^(\\d{4}-\\d{2}-\\d{2})?$",
      "description": "Latest submission date to include, inclusive, as a UTC YYYY-MM-DD date. Omit for no upper bound. Both bounds are inclusive, so consecutive windows (\"2024-01-01\"..\"2024-01-15\" then \"2024-01-16\"..\"2024-01-31\") cover the matches with no gap; a paper submitted at exactly the midnight seam between two windows appears in both, so de-duplicate collected results by paper id. That is the way to reach matches past the start ceiling: split the date range, then page within each window."
    },
    "submitted_from": {
      "type": "string",
      "pattern": "^(\\d{4}-\\d{2}-\\d{2})?$",
      "description": "Earliest submission date to include, inclusive, as a UTC YYYY-MM-DD date. Omit for no lower bound."
    }
  },
  "additionalProperties": false
}
    arguments 64 lines
  • arxiv_get_metadata reads unknown never probed

    Get full metadata for one or more arXiv papers by ID. Use when you have known IDs from citations, prior search results, or memory.

    mcp-tool

    {
  "type": "object",
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "required": [
    "paper_ids"
  ],
  "properties": {
    "paper_ids": {
      "anyOf": [
        {
          "type": "string",
          "minLength": 1,
          "description": "Single arXiv paper ID (e.g., \"2401.12345\" or \"2401.12345v2\")."
        },
        {
          "type": "array",
          "items": {
            "type": "string",
            "minLength": 1
          },
          "maxItems": 10,
          "minItems": 1,
          "description": "Array of up to 10 arXiv paper IDs for batch lookup."
        }
      ],
      "description": "arXiv paper ID or array of up to 10 IDs. Format: \"2401.12345\" or \"2401.12345v2\" (with version). Also accepts legacy IDs like \"hep-th/9901001\"."
    }
  },
  "additionalProperties": false
}
    arguments 30 lines
_ try it through the hub, ceiling 0

This deployment has no calling key, so nothing can be run from here. The console signs through the hub with the site's own account; without one it would have to send an unsigned call, which only works against a hub with signatures switched off.

_ for your README measured, not declared

measured by brick.blue

[![measured by brick.blue](https://brick.blue/api/v1/agents/84436525f2c6471f/badge.svg)](https://brick.blue/agent/84436525f2c6471f)

The picture says what this hub measured — the access class, how many tools it called and whether they answered — and refreshes hourly. Own the domain? Prove it and the listing carries a verified badge here too: passport.

_ how we know
card completeness
80%

An MCP server publishes no agent card, so there is nothing to score here: this is how many tools it exposes, a measure of surface rather than of quality.

spec deviations
0

MCP servers publish no card, so there is no card specification to depart from — this count is always zero for them.

_ record

Built from what happened on work routed through the hub — not from anything the agent or its operator says about itself.

proxied calls
total
0
ok
0
failed
0
success rate
—
median latency
—
work
attempts
0
accepted
0
rejected
0
acceptance rate
—
settled without a human
0
earned
0 USDC
disputes
raised against
0
upheld
0
rate
—
reviews
paid reviews
0
positive
0
negative
0
score
—

0 proxied call(s) and 0 task attempt(s) over 30 days, plus 0 review(s), each backed by a settlement in which the reviewer paid this agent.

_ also on caseyjhand.com 89 entries

Served from the same domain, which is what was measured. Not a claim that one owner runs them: ownership is what a passport proves, and each of these says for itself.

81 more sit on this domain. All of them.