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.

three.ws Knock

NPM · @THREE-WS/KNOCK-MCP

Reach a real person: quote their door, browse who is reachable, knock, and read the reply.

Available components

−2 this week 76 Trust /100

Last scanned (5 hours ago)

Scores higher than 57% of npm servers

Category Communication

Trust breakdown (7 categories)

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. How we score → Why this is hard to score →

Supply Chain Security81
  • No malware found by supply-chain analysis.Pass
  • CVE check failed: a known high-severity CVE affects zod 3.25.76, a direct dependency. A fixed version is available. View diagnostics → Fail
  • No install/post-install scripts declared.Pass
  • 30 of 93 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency48
  • Source repository is publicly reachable at the declared URL. View diagnostics → Pass
  • Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
  • Clear OSI-approved license (Apache-2.0).Pass
  • Actively maintained (last published 43 days ago).Pass
  • Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability69
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 1112 tokens (~185/item across 6 items; 6 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 Management83
  • Stability observed for 25 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
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
  • All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
  • An AI judge read all 7 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
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.

  • 24 Oct 26 −4
    • CVE-2023-54404 affects this package: high ▼ security
    • Known CVEs: pass → fail ▼ security
  • 22 Oct 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes.

  • 20 Oct 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

  • 18 Oct 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 57 to 60. That category is still filling its 30-day observation window: 17 days of observed history at the previous scan, 18 at this one. The score rises as the window fills, whether or not the server changes.

  • 16 Oct 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 50 to 53. That category is still filling its 30-day observation window: 15 days of observed history at the previous scan, 16 at this one. The score rises as the window fills, whether or not the server changes.

  • 14 Oct 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 43 to 47. That category is still filling its 30-day observation window: 13 days of observed history at the previous scan, 14 at this one. The score rises as the window fills, whether or not the server changes.

  • 13 Oct 26 +6
    • Stability: unverified → 0.43 ▲ functional
  • 1 Oct 26 +15
    • Malware scan: unverified → pass ▲ security
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 25 Oct 2026 · Analysed npm/@three-ws/knock-mcp@0.1.0

Provenance No attestation

The registry publishes no build provenance for this version, so there is nothing to verify.

Result No attestation
Ecosystem npm

Background: How many MCP packages publish verified provenance →

Vulnerabilities 1 finding
ID CVE Severity Vector Fix available
GHSA-mrhx-c7xc-m3w5 CVE-2023-54404 high CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H yes

Background: What a vulnerability scan can and cannot prove →

Dependencies 93 packages
Packages resolved 93
Stale 30
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 6 exposed · ~915 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
knock_act ~193

Act on one knock in your own inbox: mark it read, dismiss it, or reply to it. A reply is delivered back through the receipt link the sender already holds, so answering never hands out your email or any other address. Set block to stop hearing from that sender: they are matched on the wallet that paid when there was one, and they see the same answer a shut door gives, so they cannot tell it was them. Acting on a knock also settles its companion delivery, so your avatar will not walk on later to announce something you have already dealt with. Requires THREE_WS_API_KEY.

NameTypeReqDescription
blockboolean–Never hear from this sender again.
knock_idstringyesThe knock id from knock_inbox.
replystring–What to write back. The sender reads it through their receipt link.
statusstring–The new status. Use replied together with reply.

No output schema declared.

No examples provided.

knock_directory ~81

List every open, listed door on three.ws, cheapest first: who is reachable, for how much, and what they say they want to be reached about. This is how an agent discovers which humans it can pay to talk to. Costs nothing and needs no credential. Read-only.

NameTypeReqDescription
limitinteger–How many doors to return. Default 60.

No output schema declared.

No examples provided.

knock_inbox ~122

Read the knocks your own door has taken, newest first, each with what the sender paid for it. Use this to triage: the amount is the sender's own statement of how much reaching you was worth. Also returns totals (waiting, all time, earned). Requires THREE_WS_API_KEY for the account that owns the door; it can never read anyone else's. Read-only.

NameTypeReqDescription
beforestring–ISO timestamp cursor for paging.
limitinteger–Page size. Default 30.
statusstring–Filter to one status.

No output schema declared.

No examples provided.

knock_quote ~124

Read one person's public door on three.ws: the price of a single message to them, which chains they take it on, how long the message may be, what they say they want to hear about, and the exact endpoint to use. Costs nothing and needs no credential. ALWAYS call this before knock_send on a priced door: the owner can change their price at any time, and the quote you show a human before they approve a payment must be the live one. Read-only.

NameTypeReqDescription
handlestringyesThe recipient's three.ws username, with or without the @.

No output schema declared.

No examples provided.

knock_receipt ~102

Read the state of a knock you sent, using the receipt_url it returned. Statuses: pending (not opened), read (opened, no answer written), replied (they wrote back, and the reply is included), dismissed (read and closed). The receipt carries its own proof, so this needs no account and reveals nothing about the recipient beyond the reply they chose to write. Read-only.

NameTypeReqDescription
receipt_urlstringyesThe receipt_url a knock returned.

No output schema declared.

No examples provided.

knock_send ~293

Send exactly one message to a person through their three.ws door. The message lands in their inbox AND their 3D companion walks on screen wherever they are on the site and delivers it out loud, naming you and what you paid. On a FREE door this sends immediately. On a PRICED door this tool does not pay and cannot pay: it returns `payment_required` together with the recipient, the amount, the token, the chain and the endpoint, so a human can approve the spend and pay from their own wallet (npx @three-ws/knock, or any x402 client). Pass `request_id` to make retries safe: the same id never knocks, or charges, twice. Refusals (shut door, daily cap, message too long) happen before any payment, so a rejected knock is never a paid one.

NameTypeReqDescription
fromstringyesWho is knocking. Shown to them and spoken out loud.
messagestringyesThe message. Shown in full, never read aloud.
request_idstring–Idempotency key. A retry returns the first knock instead of knocking twice.
subjectstring–One line their companion says out loud. Make it the reason you are knocking.
tostringyesRecipient username, with or without the @.
urlstring–An http(s) link about you.

No output schema declared.

No examples provided.

Install

How do I install the three.ws Knock MCP server?

three.ws Knock runs locally as an npm package, launched with npx -y @three-ws/knock-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

npm · @three-ws/knock-mcp

# add to Claude Code
claude mcp add nirholas-knock-mcp -- npx -y @three-ws/knock-mcp
Common questions

What is the three.ws Knock MCP server?

three.ws Knock is an MCP server listed in the public MCP registry as io.github.nirholas/knock-mcp. Reach a real person: quote their door, browse who is reachable, knock, and read the reply. This page covers its npm package (@three-ws/knock-mcp).

Is the three.ws Knock MCP server safe to use?

three.ws Knock scores 76 out of 100 on VerifyMCP. We recorded 1 known advisory against it as of 25 October 2026. It declares no install or post-install scripts. 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 three.ws Knock MCP server expose?

three.ws Knock exposes 6 tools: knock_quote, knock_directory, knock_send, knock_receipt, knock_inbox, knock_act. Their descriptions and schemas cost roughly 915 tokens of context every time the server is loaded.

Is the three.ws Knock MCP server still maintained?

three.ws Knock is still listed as active in the MCP registry. We last reached this channel on 25 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What licence is the three.ws Knock MCP server under?

three.ws Knock declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.