A TypeScript implementation of a simple MCP server that exposes datetime information to agentic systems and chat REPLs
Nothing to run from here
Wixzel Voice is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
99/100
trust score
48
tools declared
local
or remote
—
downloads a day
What it does
Lets Claude, Cursor or any MCP client build voice agents and place real phone calls.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --env 'WIXZEL_API_KEY=${WIXZEL_API_KEY}' --transport stdio wixzel-voice-mcp '--' npx -y wixzel-voice-mcp
# 2 — ask your agent for something these tools do
# list_api_keys, create_api_key, rotate_api_key and 45 more
It needs WIXZEL_API_KEY: this reads it from your environment, so export it first. Remote: OAuth sign-in · Package: API key.
Wixzel Voice costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
WIAll 1 server
wixzel1 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one can run either way — at an address its publisher runs, or as a process under your own user, with your files and your network — and this marketplace is in the path of neither.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. Claude Code reads WIXZEL_API_KEY (Your Wixzel Voice API key, wv_live_... or wv_test_..., from https://voice.wixzel.com/connect) from your environment each time it starts the server, so the value stays out of its config: export it in your shell profile (macOS, Linux), or run setx WIXZEL_API_KEY … and open a new terminal (Windows). The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user --env 'WIXZEL_API_KEY=${WIXZEL_API_KEY}' --transport stdio wixzel-voice-mcp '--' npx -y wixzel-voice-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Or connect to its publisher’s address
Its publisher also runs it at https://mcp.voice.wixzel.com/mcp: nothing to start on your machine, and this marketplace is not in the path.
claude mcp add --scope user --transport http wixzel-voice-mcp https://mcp.voice.wixzel.com/mcpFree on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it asks of you
It reaches a system of yours, so it needs your own credential rather than the publisher’s: WIXZEL_API_KEY — Your Wixzel Voice API key, wv_live_... or wv_test_..., from https://voice.wixzel.com/connect. You set it in your own client’s config, on your machine — this marketplace never holds it.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
48 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
List the account's API keys with their scopes, spend limits and last use. Secrets are never returned, only a prefix and last four characters.
No parameter schema on file.
Mint a new scoped key. The secret is returned ONCE in the result and cannot be retrieved again: hand it to the user immediately and do not log it elsewhere. A key cannot grant scopes the calling key does not hold (scope_escalation). Grant only what the integration needs; billing:write authorises…
No parameter schema on file.
Issue a replacement secret. The old key keeps working for 24 hours so a deploy can roll over without a gap. The new secret is returned once; hand it to the user immediately.
No parameter schema on file.
Revoke immediately, with no grace period. Anything still using it starts failing at once, including this MCP server if it is the key in use. Not reversible.
No parameter schema on file.
List appointments booked by agents or created directly, with status scheduled, completed or canceled.
No parameter schema on file.
Fetch one appointment by id.
No parameter schema on file.
Book an appointment against a lead and agent.
No parameter schema on file.
Reschedule or annotate an appointment. Only the fields you send are changed.
No parameter schema on file.
Permanently delete an appointment. To cancel but keep the record, update its status to canceled instead. Not reversible.
No parameter schema on file.
List the MCP servers (connected apps) whose tools agents can use during calls, with each one's status and tools.
No parameter schema on file.
Fetch one MCP server: its status, last error, and every tool it offers with whether it is enabled.
No parameter schema on file.
Connect another app's remote MCP server so agents can look things up and make changes in it during calls. Its tools are read straight away; only read-only ones are enabled until you choose otherwise. For an OAuth server, follow with connect_mcp_server_oauth. Attach it to an agent with…
No parameter schema on file.
Change an MCP server, or which of its tools agents may call. Only the fields you send are changed.
No parameter schema on file.
Disconnect an MCP server. Every agent using it loses its tools. Not reversible.
No parameter schema on file.
Reconnect to the server and read its tools again. The answer's status and last_error say whether it is reachable.
No parameter schema on file.
Start signing in to an MCP server whose auth type is oauth. Give the user the authorization_url to open in their browser; once they sign in the server is connected. The link expires in ten minutes.
No parameter schema on file.
List the account's own tools (webhook and client) that agents can use during calls. Attach them with tool_ids on create_agent or update_agent.
No parameter schema on file.
Fetch one tool with its parameters and settings. Header values are never returned.
No parameter schema on file.
Create an action an agent can take during a call. type webhook: the agent's arguments go to your endpoint and its answer comes back to the agent; works on phone and web calls. type client: the caller's app or website runs it (registered in the realtime SDK's tools option, or on the agent's mascot…
No parameter schema on file.
Change a tool. Only the fields you send change; the type cannot.
No parameter schema on file.
Permanently delete a tool. Agents using it lose it. Not reversible.
No parameter schema on file.
Run a webhook tool once with sample arguments and see what the agent would hear back: the status, the time and the answer. The request is signed exactly like a call's.
No parameter schema on file.
No parameter schema on file.
List the WhatsApp business numbers connected to this account, with the agent answering each, its quality rating and whether its webhooks are arriving.
No parameter schema on file.
Fetch one connected WhatsApp number by id.
No parameter schema on file.
Connect a WhatsApp number from the user's own Meta app: its phone_number_id and waba_id (WhatsApp > API Setup), a permanent access token and the app secret. The easier path for most people is the console, where Connect WhatsApp opens Meta's own sign-up with no app needed; suggest that first. The…
No parameter schema on file.
Change which agent answers a WhatsApp number (inbound_agent_id, or null for none), its label, or how the agent behaves on it. Only the fields you send change.
No parameter schema on file.
Disconnect a WhatsApp number. Its chats, contacts and scheduled messages kept in Wixzel Voice are deleted; the number stays the user's in WhatsApp Manager. Not reversible: confirm with the user first.
No parameter schema on file.
Read a WhatsApp number's name, quality rating and messaging limit from Meta again.
No parameter schema on file.
List a WhatsApp number's message templates, as last synced. Templates are written and approved in WhatsApp Manager; only APPROVED ones can be sent, and they are the only messages allowed more than 24 hours after a person last wrote.
No parameter schema on file.
Copy a WhatsApp number's templates from WhatsApp Manager now, after the user has created or edited one there.
No parameter schema on file.
Write a WhatsApp message template and submit it to Meta for review. It is PENDING until Meta approves it, usually within minutes; only approved templates can be sent. Number the values in the body {{1}}, {{2}}, ... and give one example value for each, in order. Use UTILITY for confirmations,…
No parameter schema on file.
Delete one WhatsApp template from WhatsApp Manager and from Wixzel Voice. Agents that use it stop finding it. Confirm with the user first.
No parameter schema on file.
Send one WhatsApp message from a connected number: text, a knowledge-base file (knowledge_file_id), or an approved template. Exactly one of the three. Text and files only work within 24 hours of the person's last message to that number; otherwise send a template. Set send_at to schedule it. This…
No parameter schema on file.
List the WhatsApp chats agents have had, newest first, optionally for one number, one agent, or open or closed ones.
No parameter schema on file.
Fetch one WhatsApp conversation with its messages, oldest first, including voice-note transcripts and what it cost.
No parameter schema on file.
List reminders, follow-ups and other WhatsApp messages scheduled to send later.
No parameter schema on file.
Cancel a scheduled WhatsApp message that has not been sent yet.
No parameter schema on file.
List the voice agents on this account. Cursor-paginated: pass next_cursor as starting_after for the next page.
No parameter schema on file.
Fetch one agent by id, including its full voice configuration, prompt and mascot.
No parameter schema on file.
Create a voice agent. Pick models with list_engines first: an unavailable or mistyped model is rejected with unsupported_model. voice is either a composed pipeline {stt, llm, tts} or a single {realtime} model, never both; an openai/ realtime model (GPT-Live) may also carry {llm} for the model that…
No parameter schema on file.
Build an agent on the server from the business's website_url (its pages become the agent's knowledge base, its colour and tone the mascot), a description, or both, and the role it plays, e.g. "receptionist". Takes up to a minute; this tool waits for it and returns the finished agent (mascot.status…
No parameter schema on file.
Ring a number and have the agent speak one phrase, then hang up. A probe, not a conversation: it answers "is the trunk configured and does this agent sound right" in seconds. For a real conversation use place_call. THIS RINGS A REAL PHONE AND SPENDS REAL CREDIT — the meter runs on a test call…
No parameter schema on file.
Change an agent. Only the fields you send are changed; omit the rest. Sending voice replaces the whole voice object.
No parameter schema on file.
Replace an agent's mascot spec or change its welcome message. A spec replaces the old one whole, so send the full spec with your changes in it. Sharing the mascot publicly (which lets strangers call the agent on the account's credit) is done in the console, not here. For a new look made on the…
No parameter schema on file.
Give an agent's mascot a new look made on the server, from a website_url, a description, or both; neither for a random one. The agent's prompt is not changed. Takes up to a minute; this tool waits for it and returns the agent (mascot.status ready, or failed with mascot.error and the old look).…
No parameter schema on file.
Give the agent's mascot a new public_id. Its old share link and website embed code stop working at once. Also needs calls:write.
No parameter schema on file.
Permanently delete an agent and its mascot. Phone numbers and campaigns pointing at it stop working. Not reversible.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Wixzel Voice or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATrust score 99 / 100
Release 0.12.0 · scanned 2 Nov 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score99
Findings0 threat · 0 capability
Capabilityminimal
Coveragesource
Verificationrepo
Surface digestd0bd60f3efee3c350e5d211ab3aa124e98abeaa82a433a6d35ddcaceb7c87952
Score vectorthreat 100 → client 99
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | minimal | capability blast radius (minimal) — client exposure if the model is manipulated | −0 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfacenot available
Implementation sourceread
Package metadataread
Live transportnot available
No tools were enumerated, so prompt-injection, capability and toxic-flow analysis had no tool surface to inspect. To grade a package’s real runtime tools, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. minimal is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| No per-tool rows for this release — see the note below. | ||
No tool surface was enumerated, so there is nothing to attribute a capability to.
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digestd0bd60f3efee3c35
Scanned2026-11-02
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 0.12.0. Pin a release below and the command asks for exactly that one.
0.12.0
undated
latestNo release note was published with this version.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Wixzel Voice can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.