# io.github.arxivsub/arxivsub-mcp (npm · arxivsub-mcp)

Search arXiv and major AI/ML/CV/NLP/robotics conference papers and their related work.

- Trust score: 67/100 (medium)
- Change this week: +24
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- npm · `arxivsub-mcp`: 67/100 (this document), [markdown](https://verifymcp.io/servers/arxivsub-arxivsub-mcp/arxivsub-mcp.md), [page](https://verifymcp.io/servers/arxivsub-arxivsub-mcp/arxivsub-mcp)

## Channel facts

- Registry: `npm`
- Package: `arxivsub-mcp`
- Version: `0.1.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-08-03.

- **Supply Chain Security**: 87/100
  - No malware found by supply-chain analysis.
  - Only part of the dependency tree could be resolved (95 of 99), so this covers what we could see, not the whole tree.
  - No install/post-install scripts declared.
  - Only part of the dependency tree could be resolved (95 of 99), so this covers what we could see, not the whole tree.
- **Provenance & Transparency**: 45/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 24 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 63/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 1180 tokens (~236/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 27/100
  - Stability observed for 8 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add arxivsub-arxivsub-mcp -- npx -y arxivsub-mcp
```

### Codex

```bash
codex mcp add arxivsub-arxivsub-mcp -- npx -y arxivsub-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "arxivsub-arxivsub-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "arxivsub-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add arxivsub-arxivsub-mcp --command npx --arg -y --arg arxivsub-mcp
```

### Hermes

```yaml
mcp_servers:
  arxivsub-arxivsub-mcp:
    command: "npx"
    args: ["-y", "arxivsub-mcp"]
```

### Other

```json
{
  "mcpServers": {
    "arxivsub-arxivsub-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "arxivsub-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-08-03 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-02 (score 66, +10)

- [security regression] Install scripts: pass → unverified
- [security regression] Provenance: fail → unverified
- [security improvement] Known CVEs: unverified → partial
- [security] Stability: Stability not yet verified: we do not have a sandbox capture of the MCP schema this version of the package serves yet.
- [functional regression] Capabilities: pass → unverified
- [functional regression] Maintenance: pass → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional regression] License: pass → unverified
- [functional improvement] Stability: unverified → 0.23
- [functional improvement] Dependency health: unverified → partial
- [functional] Licence: MIT

### 2026-08-01 (score 56, −10)

- [security regression] Known CVEs: partial → unverified
- [functional regression] Dependency health: partial → unverified

### 2026-07-31 (score 66, +33)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-30 (score 33, −43)

- [security regression] Known CVEs: partial → unverified
- [functional regression] Dependency health: partial → unverified
- [functional regression] Tool coverage: 100 → unverified

### 2026-07-29 (score 76, +51)

- [security regression] Provenance: unverified → fail
- [security improvement] Known CVEs: unverified → partial
- [security improvement] Install scripts: unverified → pass
- [functional improvement] License: unverified → pass
- [functional improvement] Tool coverage: unverified → 100
- [functional improvement] Schema quality: unverified → excellent
- [functional improvement] Maintenance: unverified → pass
- [functional] Licence: MIT

### 2026-07-28 (score 25, −18)

- [functional regression] Tool coverage: 100 → unverified
- [functional improvement] Dependency health: unverified → partial
- [functional] First check of Schema quality: unverified

### 2026-07-27 (score 43)

First indexed and scored.

## MCP tools (5)

### `search_papers` (~345 tokens)

Search academic papers from arXiv and major AI/CV/ML/robotics/NLP conferences via arXivSub (semantic + keyword search over a structured, summarized corpus). Use for 'find papers on X', 'latest research about Y', 'recent conference work on Z', or to start a literature review. Returns each paper with a 6-part summary (what it's about, innovations, techniques, datasets, results, limitations), authors + affiliations, pdf_url, and github_url when code is available.

Input parameters:

- `arxiv_days` (integer): Only arXiv papers from the last N days. Default 30.
- `conference_years` (array): Conference publication years to include, e.g. [2024, 2025]. Default: last two years.
- `has_code` (boolean): If true, only return papers that ship a public code repository (a reproducibility signal).
- `language` (string): Summary language, 'en' or 'zh'. Default 'en'.
- `limit` (integer): Max papers per source (server caps at 100). Default 10.
- `locations` (array): Venues to search (CASE-SENSITIVE): arxiv, CVPR, ICCV, ECCV, ICLR, ICML, NeurIPS, AAAI, MICCAI, CoRL, RSS, ACL, EMNLP, ICRA, IROS, IJCAI, SIGGRAPH, 'SIGGRAPH Asia'. Include 'arxiv' for arXiv preprints…
- `query` (string, required): Natural-language search query, e.g. 'LLM safety alignment'.

### `find_similar_papers` (~205 tokens)

Given ONE paper you already have (from a prior search result), return the most similar papers from the same corpus — semantic nearest neighbours, each with the full summary and code link. Use for 'more like this', 'related work', 'what else is in this line of work', or to expand a literature review. You can chain it: search -> pick a seed -> find_similar_papers -> repeat on a strong neighbour.

Input parameters:

- `item_id` (string, required): The seed paper's arXivSub id (uuid) from a prior search result. Never invent one.
- `item_type` (string, required): The seed paper's type, from its search 'source': 'arxiv' -> 'paper', 'conferences' -> 'ciiina'.
- `language` (string): Summary language, 'en' or 'zh'. Default 'en'.
- `limit` (integer): How many neighbours to return (server caps at 20). Default 8.

### `research_insight` (~326 tokens)

Build aggregate pivot statistics over the arXivSub corpora — the same engine as the website's Custom Chart. Use for trends, rankings, breakdowns and comparisons: 'how has interest in diffusion models trended', 'top institutions in RL', 'papers per month by keyword', 'avg citations by affiliation'. Returns rows of {dim, breakdown, value} for you to interpret and visualize. Requires Pro or active trial days.

Input parameters:

- `breakdown` (string|null): Optional second split, same value set as dim. e.g. dim=time:month, breakdown=keyword.
- `corpus` (string): 'arxiv' (recent arXiv) or 'ciiina' (conferences; 'conference' is an alias). Default 'arxiv'.
- `dim` (string, required): REQUIRED axis. arxiv: time:day|week|month|year, keyword, category, affiliation, author, citation_bin. ciiina: time:year, keyword, conference, award, affiliation, author, citation_bin.
- `filters` (object): Optional object: search, start_date, end_date, keywords[], categories[] (arxiv), conferences[]/publish_years[]/awards[] (ciiina), affiliations[].
- `measure` (string): count_papers (default) | count_authors | count_affiliations | count_keywords | avg_citation | median_citation | award_rate (ciiina only).
- `options` (object): Optional object: top_n_dim, top_n_breakdown, min_count, show_others, sort.

### `library_save` (~184 tokens)

Save (or update) a paper in the user's personal arXivSub library. Use for 'save this', 'bookmark that paper', 'add it to my <X> collection', 'note that ...', 'mark as read'. Idempotent: saving the same paper again updates only the fields you pass and never duplicates. Cannot delete — removals are done on the website.

Input parameters:

- `collection_name` (string): Optional collection/folder name; created automatically if it doesn't exist.
- `item_id` (string, required): The item's arXivSub id (uuid) from a prior search result.
- `item_type` (string, required): From the search 'source': 'arxiv' -> 'paper', 'conferences' -> 'ciiina'.
- `note` (string): Optional note (<=2000 chars).
- `reading_status` (string): Optional reading status.

### `library_list` (~120 tokens)

List the user's personal arXivSub library — collections and saved items, optionally filtered. Use for 'what have I saved', 'show my library', 'which papers are in my <name> collection', 'what am I reading'.

Input parameters:

- `collection_name` (string): Only items in this collection.
- `item_type` (string): Only papers ('paper') or conference items ('ciiina').
- `limit` (integer): Max items (server caps at 200). Default 100.
- `status` (string): Only items with this reading status.

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/arxivsub-arxivsub-mcp/arxivsub-mcp#diagnostics

## Score history

- 2026-08-03: 67
- 2026-08-02: 66
- 2026-08-01: 56
- 2026-07-31: 66
- 2026-07-30: 33
- 2026-07-29: 76
- 2026-07-28: 25
- 2026-07-27: 43

## Links

- npm package: https://www.npmjs.com/package/arxivsub-mcp
- Socket report: https://socket.dev/npm/package/arxivsub-mcp
- Repository: https://github.com/arxivsub/arxivsub-mcp
- Changelog RSS feed: https://verifymcp.io/servers/arxivsub-arxivsub-mcp/arxivsub-mcp/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/arxivsub-arxivsub-mcp/arxivsub-mcp/changelog.json
- HTML version of this page: https://verifymcp.io/servers/arxivsub-arxivsub-mcp/arxivsub-mcp