Skills/npm/hf-mcp-server

hf-mcp-server@0.4.37

CAUTION· not started

Publisher claims: "Official Hugging Face MCP Server" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher llmindset · publishing for 514 daysInstalls 2,928Last detonated 18 h ago

Coverage

Not started: dependency install failed; the server itself was not observed.
Tested
the dependency install step · every file read, file written and host contacted by what ran
Not tested
server startup · the tool list · tool behavior

What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

It will ask for
DEFAULT_HF_TOKENLOGGING_HF_TOKENMETRICS_PAGE_PASSWORDConfiguration, not credentials: CORS_ALLOWED_ORIGINS, GRADIO_DISCOVERY_CONCURRENCY, GRADIO_SCHEMA_CACHE_TTL, GRADIO_SCHEMA_TIMEOUT, GRADIO_SPACE_CACHE_TTL, GRADIO_SPACE_INFO_TIMEOUT, HSTS, LOG_QUERY_EVENTS, LOG_SYSTEM_EVENTS, LOGGING_DATASET_ID, MCP_ALLOWED_HOSTS, SHUTDOWN_GRACE_MS, TRANSPORT, USER_CONFIG_API, VITEST, WEB_APP_PORT
Where it talks to
no vendor host named in the codeAlso huggingface.co, hf.co, react.dev — not the publisher's domain; check what for before connecting.
What we saw when it ran
no credential files read · no network
What we could not test
server startup · the tool list · tool behavior

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names huggingface.co, hf.co, react.dev as well as its own vendor. The guard injects DEFAULT_HF_TOKEN, LOGGING_HF_TOKEN, METRICS_PAGE_PASSWORD only into requests to the host you bind (and its subdomains) and refuses the rest.

$vet-skills run --net declared -- npx -y @llmindset/hf-mcp-server

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

What changed since 0.4.34

vet-skills diff hf-mcp-server 0.4.34 0.4.37

In the sandbox, this version reached for nothing 0.4.34 did not. Compared from the two recorded runs, not from source.

0.4.34 was not run in the sandbox (install failed); the comparison covers what static analysis recorded only.

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; not started: dependency install failed; the server itself was not observed) on 2026-11-19 and scored it 78/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: URL with embedded credentials; Base64 encoding combined with network call. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 21 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
20sprocessdependency install exit 1 · (Use `node --trace-warnings ...` to show where the warning was created) npm error Cannot read properties of null (reading 'edgesOut') npm error A complete log of this run can be found in: /tmp/ks-detonate-DZKgsf/task/home/.npm/_logs/2026-11-19T06_32_43_471Z-debug-0.log
20sclaimednot started: dependency install failed (exit 1); the server itself was not observed
20sprocessscript backend runs setup only
20sprocessno install scripts to re-run

Static findings

MEDIUM
MCP server using network transport (SSE/HTTP)package.json:9 · MCP server using network transport (SSE/HTTP) — data sent over network"hf-mcp-server-http": "./dist/server/streamableHttp.js",Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
MEDIUM
File type combined with upload actiondist/web/assets/main-BJNEMfD1.js:52 · File type combined with upload action — agent may exfiltrate local data files…>fetch(e).then(e=>{if(!e.ok)throw Error(`Failed to fetch: ${e.status}`);return e.json()});function ih(){let{data:e,error:t}=tt(`/api/transport-metrics`,rh,{refreshInterval:3e3,revalidateOnFocus:!0,revalidateOnReconnect:!0});return t?(0,z.j…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
HIGH
URL with embedded credentialsdist/web/assets/main-Ct5nq0rL.css:1 · URL with embedded credentials — may be phishing or credential leaking@import "https://fonts.googleapis.com/css2?family=Source+Sans+Pro:ital,wght@0,400;0,600;0,700;0,900;1,400;1,700&display=swap";@layer components;@layer prope…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST07
HIGH
Base64 encoding combined with network calldist/web/assets/main-BJNEMfD1.js:9 · Base64 encoding combined with network call — potential data exfiltration…~`&&n!==`$!`||e++}n=r}while(n)}function Ep(e,t,n){if(t=CSS.escape(t)===t?t:`r-`+btoa(t).replace(/=/g,``),e.style.viewTransitionName=t,n!=null&&(e.style.viewTransitionClass=n),n=getComputedStyle(e),n.display===`inline`){if(t=e.getClientRect…Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST07
MEDIUM
MCP server using network transport (SSE/HTTP)package.json:10 · MCP server using network transport (SSE/HTTP) — data sent over network"hf-mcp-server-http-json": "./dist/server/streamableHttp.js"Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
MEDIUM
File type combined with upload actiondist/web/assets/main-BJNEMfD1.js:8 · File type combined with upload action — agent may exfiltrate local data files…nstable_NormalPriority,Ye=t.unstable_LowPriority,Xe=t.unstable_IdlePriority,Ze=t.log,Qe=t.unstable_setDisableYieldValue,$e=null,et=null;function tt(e){if(typeof Ze==`function`&&Qe(e),et&&typeof et.setStrictMode==`function`)try{et.setStrict…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File type combined with upload actiondist/web/assets/main-BJNEMfD1.js:9 · File type combined with upload action — agent may exfiltrate local data files…(!t.ok)throw Error(`Failed to fetch Skills metrics: ${t.status}`);return await t.json()}function ot(e){var t,n,r=``;if(typeof e==`string`||typeof e==`number`)r+=e;else if(typeof e==`object`){if(Array.isArray(e)){var i=e.length;for(t=0;t<i;…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07

Cost to the agent

Instructions loaded
≈48 tok
1 file · 81 in bundle
Tool list per prompt
≈0 tok
0 tools · 0 per tool
Handshake
no reply
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 0

Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.

Known vulnerabilities

No advisories in OSV for @llmindset/hf-mcp-server@0.4.37 as of 18 h ago. This is the package itself, not its dependencies.

What would make this Verified

2 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim llmindset: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Obfuscated code≈ +5 on the totalRemove base64-assembled strings, eval of decoded text and minified blobs. Reviewers and scanners both treat unreadable code as hostile.
  3. URL with embedded credentials≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  4. Base64 encoding combined with network call≈ +4 on the totalPut hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
  5. MCP server using network transport (SSE/HTTP)≈ +2 on the totalNamespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
  6. File type combined with upload action≈ +2 on the totalRemove the call, or declare the host in frontmatter and explain in the README what is sent and why.
Publish a new version to npm, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim llmindsetPropose a patch to the maintainer