Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

Hands for Agents

REMOTE · MCP.HANDSFORAGENTS.COM · 2 COMPONENTS

Human-operated physical world in the EU: make, assemble, measure, verify, ship.

0 this week 74 Trust /100

Last scanned (20 hours ago)

Scores higher than 61% of remote servers

Reachable. Last reached 21 Nov 2026.

Category IoT and Hardware

Recent critical change

Authorization (2 Oct 2026). See the changelog before you install this server.

Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →

Endpoint Security51
Transport & Reachability100
Schema Quality & AI Usability64
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 2064 tokens (~412/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
  • No destabilizing schema changes in the last 30 days.Pass
Tool Coverage100
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 100% of tool parameters carry a description.Pass
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
  • An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
What we tested

Checks we ran (12)

  • TLS certificate
  • Authorization
  • HSTS header
  • DNSSEC
  • Transport
  • Schema quality
  • Stability
  • Tool coverage
  • Injection markers
  • Destructive annotations
  • Judged manipulation
  • MCP protocol

Checks we could not complete (1)

These are gaps on our side, not findings about this hosted endpoint. We only credit what we can confirm, so a category made up entirely of them scores 0.

  • HTTPS in Endpoint SecurityHTTPS enforcement could not be verified: the plaintext port answered with HTTP 406, which proves neither a plaintext path nor enforcement.

Does not apply to this hosted endpoint (2)

  • Supply Chain SecurityScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.
  • Provenance & TransparencyScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.

The score does not measure everything. What the score does not measure →

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 11 Nov 26 0
    • Stability: 0.97 → pass security
  • 10 Nov 26 0
    • Stability: pass → 0.97 functional
  • 9 Nov 26 0
    • Stability: 0.97 → pass security
  • 8 Nov 26 0
    • Stability: pass → 0.97 functional
  • 6 Nov 26 0
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
  • 29 Oct 26 +2
    • Stability: fail → pass ▲ security
  • 3 Oct 26 0
    • The server rewrote its instructions, which are the text every model session reads security
    • Tool “list_services” rewrote its description, which is the text the model reads security
    • Tool “request_quote” rewrote its description, which is the text the model reads security
    • Tool “confirm_delivery” rewrote its description, which is the text the model reads security
    • Tool “create_task” rewrote its description, which is the text the model reads security
    • Tool “get_status” rewrote its description, which is the text the model reads security
    • Tool coverage: 89% → 100% ▲ functional
    • “request_quote” reworded the description of “vat_id” cosmetic
    • “get_status” reworded the description of “access_token” cosmetic
    • “get_status” reworded the description of “full” cosmetic
    • “get_status” reworded the description of “message” cosmetic
    • “list_services” reworded the description of “full” cosmetic
    • “request_quote” reworded the description of “allow_anonymised_example” cosmetic
    • “request_quote” reworded the description of “country” cosmetic
    • “request_quote” reworded the description of “deliverables” cosmetic
    • “request_quote” reworded the description of “delivery_address” cosmetic
    • “request_quote” reworded the description of “description” cosmetic
    • “request_quote” reworded the description of “email” cosmetic
    • “request_quote” reworded the description of “name” cosmetic
    • “request_quote” reworded the description of “services” cosmetic
    • “confirm_delivery” reworded the description of “accepted” cosmetic
    • “confirm_delivery” reworded the description of “access_token” cosmetic
    • “confirm_delivery” reworded the description of “task_id” cosmetic
    • “create_task” reworded the description of “access_token” cosmetic
    • “create_task” reworded the description of “allow_anonymised_example” cosmetic
    • “create_task” reworded the description of “quote_id” cosmetic
  • 2 Oct 26 0
    • Authorization: partial → fail ▼ critical
    • Tool “confirm_delivery” is now declared destructive security
    • Tool “confirm_delivery” changed its title: Confirm or dispute delivery cosmetic
    • Tool “create_task” changed its title: Accept quote, get payment link cosmetic
    • Tool “get_status” changed its title: Get status or answer operator cosmetic
    • Tool “list_services” changed its title: List services cosmetic
    • Tool “request_quote” changed its title: Request a quote cosmetic
Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 21 Nov 2026 · Probed https://mcp.handsforagents.com/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=handsforagents.com CN=YE2,O=Let's Encrypt,C=US 23 Sept 2026 22 Dec 2026 ECDSA 256 ECDSA-SHA384 63c81a467d4a5d4d32335208baaaa216292
SANs: *.handsforagents.com, handsforagents.com
CN=YE2,O=Let's Encrypt,C=US (CA) CN=Root YE,O=ISRG,C=US 3 Sept 2025 2 Sept 2028 ECDSA 384 ECDSA-SHA384 4df3b15dd6c0784c507cd37b58e6f115
CN=Root YE,O=ISRG,C=US (CA) CN=ISRG Root X2,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 ECDSA 384 ECDSA-SHA384 872165fc34b6e5fba8add5b3705fb53a
CN=ISRG Root X2,O=Internet Security Research Group,C=US (CA) CN=ISRG Root X1,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 ECDSA 384 SHA256-RSA 6c8f1dc727c7117f7baf853ac980f9cd

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of mcp.handsforagents.com. — Not signed

Zone DS Keys Algorithms Outcome
. Trust anchor 20326, 38696 8, 8 Verified
com. DS published 19718 13 Verified
handsforagents.com. No DS Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication No authorisation required

The endpoint answered without asking for a token. Anyone who knows the URL can reach it.

Result No authorisation required
HTTP status 200
Header Value
strict-transport-security max-age=31536000
x-content-type-options nosniff

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://mcp.handsforagents.com/mcp Verified 200
http (plaintext) http://mcp.handsforagents.com/mcp Inconclusive 406
MCP tools · 5 exposed · ~1,430 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
confirm_delivery ~163

Accepts or disputes the delivery of a task. accepted=true closes the task and, for a card hold, captures it. accepted=false with a reason opens a dispute, answered by a human within 2 business days; it is possible at the evidence stage, before the balance is paid, as well as after delivery. Without either, the task counts as accepted 7 days after delivery. Returns the task status.

NameTypeReqDescription
acceptedbooleanyestrue accepts the delivery, false disputes it.
access_tokenstringyesAccess token of the quote and its task, issued with the quote.
reasonstring–Required when accepted is false: what differs from the quote.
task_idstringyesId of the task, of the form t_...

No output schema declared.

No examples provided.

create_task ~283

Accepts a quote and the terms of service (quote_id, access_token, terms_accepted=true, terms_version as stated in the quote) and creates the task. Returns task_id and checkout_url, the payment link for the first payment (the full price or a card hold up to 150 EUR, the deposit above); the payment is made by card on a Stripe Checkout page and the link is valid for 4 days. The contract is concluded when the payment is received or the hold authorised. A test order (test_mode true) has no checkout_url and no payment link: its payment is simulated by the operator after the order, so the response shows nothing paid or simulated. A repeated call for the same quote returns the same task.

NameTypeReqDescription
access_tokenstringyesAccess token of the quote and its task, issued with the quote.
allow_anonymised_exampleboolean–Optional. Consent to publish this task as an anonymised example (terms, section 10); true only when the client explicitly agreed. Default false.
quote_idstringyesId of the quote, of the form q_...
terms_acceptedbooleanyesAcceptance of https://handsforagents.com/terms.html by the operator.
terms_versionstringyesVersion of the terms being accepted, currently 2026-09-29.

No output schema declared.

No examples provided.

get_status ~269

Returns the current status of a quote (id = quote_id) or a task (id = task_id): status, next_action (a short text on the next step), payments, updates, and messages with the operator (newest first, from operator or you). The answer is short: older_updates and older_messages count what is left out; full=true returns everything. With the optional message the call also stores a plain-text reply to the operator and e-mails it to the operator; message_sent is then true. While the operator has a message that has not been returned yet, the reply is not stored (message_sent: false) and the answer contains that message. The same text again within 10 minutes is ignored (duplicate_ignored: true). A status can change at any time; every call returns the state at that moment. test_mode true marks a test order: nothing is paid or invoiced.

NameTypeReqDescription
access_tokenstringyesAccess token of the quote and its task, issued with the quote.
fullboolean–true returns the complete answer.
idstringyesquote_id or task_id.
messagestring–Optional reply to the operator: plain text, 1-2000 characters. Without it the call is a plain status check.

No output schema declared.

No examples provided.

list_services ~92

Returns what the service does and does not do, the refused categories, prices and how to pay, as a short summary. No authentication. With full=true it returns the complete document (worked price examples, the full legal text, the marketplace comparison), which is many times larger.

NameTypeReqDescription
fullboolean–true returns the complete document (the same as /services.json); false (default) returns a short summary.

No output schema declared.

No examples provided.

request_quote ~623

Creates a request for a fixed-price quote from a human. The request is not binding and nothing is charged. Required: description, email and name (the person or company on the invoice, normally the recipient or company in delivery_address). delivery_address is needed when a physical item is shipped, handed over or collected: one plain string with recipient, street, postcode and city, country; a phone number is optional. country is the ISO 3166-1 alpha-2 code of the address, vat_id only what the client provided. The fields are sent flat, as in the examples. Returns quote_id (of the form q_..., issued only in this response), access_token (shown once; the only key to the quote and its task), nda_url and upload_url (the address input files are POSTed to). A human answers within 24 hours, every day, with a quote or a refusal. No authentication. Shortest possible request: {"description":"Make 20 rubber gaskets, 60 mm outer diameter, black EPDM.","email":"ops@example.com","name":"Example Inc."}. With a delivery address and an input file: {"description":"Make 20 rubber gaskets, 60 mm outer diameter, black EPDM.","email":"ops@example.com","name":"Example Inc.","delivery_address":"Example GmbH, Musterstr. 1, 10115 Berlin, DE, +49 30 000000","input_files":["https://example.com/gasket-drawing.pdf"]}

NameTypeReqDescription
agent_namestring–Name of the agent or platform. Optional.
allow_anonymised_exampleboolean–Consent to publish this task as an anonymised example; true only when the client explicitly agreed. Default false.
countrystring–Operator's country, ISO 3166-1 alpha-2, upper case, normally that of the delivery address. Optional.
currencystring–EUR or USD. Default EUR.
deadlinestring–Latest acceptable delivery date, ISO 8601. Optional.
deliverablesstring–What the client expects to receive. Optional; otherwise part of description.
delivery_addressstring–Required for a physical shipment. One plain string: recipient, street, postcode and city, country, as given by the client; a phone number last is optional.
descriptionstringyesWhat is to be done, plainly: quantity, material, colour, size, tolerances, what counts as done.
emailstringyesContact e-mail of the operator; the quote is sent here.
input_filesarray–HTTPS URLs of input files (photos, sketch, STEP/STL model, PDF drawing). At most 20.
namestringyesRequired. The person or company on the invoice, normally the recipient or company in delivery_address.
servicesarray–Service ids. Optional.
vat_idstring–VAT or tax id for the invoice, as given by the client. Optional.

No output schema declared.

No examples provided.

Install

How do I install the Hands for Agents MCP server?

Hands for Agents is a hosted endpoint at https://mcp.handsforagents.com/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

remote · mcp.handsforagents.com

# add to Claude Code
claude mcp add --transport http com-handsforagents-hands 'https://mcp.handsforagents.com/mcp'
Common questions

What is the Hands for Agents MCP server?

Hands for Agents is an MCP server listed in the public MCP registry as com.handsforagents/hands. Human-operated physical world in the EU: make, assemble, measure, verify, ship. This page covers its hosted endpoint (https://mcp.handsforagents.com/mcp).

Is the Hands for Agents MCP server safe to use?

Hands for Agents scores 74 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the Hands for Agents MCP server expose?

Hands for Agents exposes 5 tools: list_services, request_quote, create_task, get_status, confirm_delivery. Their descriptions and schemas cost roughly 1,430 tokens of context every time the server is loaded.

Does the Hands for Agents MCP server require authentication?

No. We connected to Hands for Agents without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

Is the Hands for Agents MCP server still maintained?

Hands for Agents is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What did VerifyMCP test on the Hands for Agents MCP server?

VerifyMCP ran 12 checks on Hands for Agents, covering TLS certificate, Authorization, HSTS header, DNSSEC and 8 more. One item could not be completed, a gap on our side rather than a finding about Hands for Agents: HTTPS. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.