MobusBLOCK
Search, preview, and analyze datasets from 20+ platforms via a single MCP connector. Works with Claude instantly.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Dataset search for AI assistants Discover, preview, and analyze datasets across 20 platforms from a single conversation.
[](LICENSE) [](https://nodejs.org) [](https://modelcontextprotocol.io) [](https://www.typescriptlang.org)
Connect to Claude
Add Mobus to Claude in under a minute. No install, no API keys, nothing to run.
- Open claude.ai (or Claude Desktop / Mobile)
- Go to Settings from the bottom left → Connectors
- Click Add custom connector
- Name it Mobus and paste this URL:
https://mcp.mobus.ai/mcp
- Start a new chat and try:
"Search for air quality datasets with a commercial license"
That's it. All 15 tools are available immediately.
What it does
Just ask your AI assistant.
"Search for air-quality datasets with a commercial license" "Preview the first 20 rows of that Zenodo dataset" "Find SEC filings mentioning climate risk" "Generate an APA citation for that Hugging Face dataset" "Check if this dataset can be used commercially" "Visualize that dataset"
Mobus fans requests out to every configured platform in parallel, checks licenses, previews data, generates citations, and traces academic lineage — failing gracefully whenever an API key is missing.
Tools
Discovery
- `s
8340a17af00cOBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add mobus --env DATAVERSE_API_TOKEN=${DATAVERSE_API_TOKEN} --env GITHUB_TOKEN=${GITHUB_TOKEN} --env GOOGLE_API_KEY=${GOOGLE_API_KEY} --env HF_TOKEN=${HF_TOKEN} -- npx -y [email protected]{
"mcpServers": {
"mobus": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"DATAVERSE_API_TOKEN": "${DATAVERSE_API_TOKEN}",
"GITHUB_TOKEN": "${GITHUB_TOKEN}",
"GOOGLE_API_KEY": "${GOOGLE_API_KEY}",
"HF_TOKEN": "${HF_TOKEN}"
}
}
}
}Exposed tools (13)
12 read · 0 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
assess_quality | read | Assess data quality: missing values, duplicates, basic statistics, and date freshness |
check_compatibility | read | Check if a dataset |
check_license | read | Check if a dataset |
compare_datasets | read | Compare 2-5 datasets side by side: columns, sizes, licenses, and column overlap |
enrich_result | read | Enrich a dataset or paper by DOI: resolve open-access links, PDF URLs, citation counts, and reference graphs via Unpaywall, OpenCitations, and NIH iCite |
find_research_datasets | read | Find datasets used in academic research for a given topic — searches Semantic Scholar and arXiv for paper metadata, and cross-references against all other sources |
find_similar | read | Find datasets similar to a given dataset based on tags, name, and description |
generate_citation | read | Generate a formatted citation (BibTeX, APA, or Chicago) for a dataset |
get_dataset_details | read | Get detailed metadata for a specific dataset (columns, row count, files, popularity) |
preview_dataset | read | Preview the first N rows of a dataset to inspect its data before downloading |
search_datasets | read | |
visualize_dataset | read | Generate an interactive ECharts dashboard to explore a dataset visually — charts, filters, table view, and export. In remote mode returns a URL; locally opens in the browser. |
watch_query | destructive | Save, remove, list, or check dataset search watches for monitoring new datasets |
Trust audit
BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (5 observation(s))
- Shell
- declared (1 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (3)
exec(`${cmd} "${filePath}"`);watch_query
@modelcontextprotocol/sdk, dotenv, pg, zod, @types/node, @types/pg, tsx, typescript
Gates applied: no_behavioural_pass.
8340a17af00cfull audit observations/trust-audit/mcp-server/mobus-ai__mobus.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 8340a17af00c | BLOCK | D | 69 | first audit |
Questions
What is the Mobus MCP server?
Search, preview, and analyze datasets from 20+ platforms via a single MCP connector. Works with Claude instantly.
What tools does Mobus expose?
13 in total: 12 read-only, 0 that write, and 1 that can delete or overwrite (watch_query). Every one is listed on this page with its risk.
Is Mobus safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (69/100) and found 1 critical or high issue in the source. Each one is listed on this page with the file and line it is on. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Mobus need?
It reads DATAVERSE_API_TOKEN, GITHUB_TOKEN, GOOGLE_API_KEY, HF_TOKEN, KAGGLE_KEY and SOCRATA_APP_TOKEN from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Mobus run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as mobus at 2.1.0.
How current is this page?
The grade is for one exact copy of the source (8340a17af00c), read on 2026-10-08. The repository is watched and re-audited when it changes.