Vapi MCP Server
tunnel 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.
What it does
Local HTTPS proxy and mock server for app tests: record, replay, mock and break API calls
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user e2e-next-tunnel-mcp '--' npx -y tunnel-use mcp
# 2 — ask your agent for something these tools do
# status, guide, list_mocks and 13 more
tunnel costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts tunnel as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
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. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user e2e-next-tunnel-mcp '--' npx -y tunnel-use mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
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.
16 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.
Whether tunnel is up and how it is set right now: version, the three directories, how many rules and logged requests it holds, the session in hand, whether recording is on, the proxy address a device points at, the CA fingerprint, and what every session on disk holds. Make this the first call.
No parameter schema on file.
tunnel's AGENTS.md, the copy this version ships: the handler(ctx) API a script rule is written against, how a request key is matched, how replay picks a recording, and the steps of a test run. Read it before writing rules. Needs no running server.
No parameter schema on file.
Every rule as it stands, whole, in the shape apply_scenario takes back — so this is also how a set of rules is exported. A key narrows it to one.
No parameter schema on file.
Configure what one endpoint answers; only the fields passed change. Give exactly one source — json_response, json_file, script, script_file, target or replay — and it settles the mode. Everything else shapes whatever that mode answers.
No parameter schema on file.
Delete one rule. The request log is left alone.
No parameter schema on file.
Apply a set of rules in one call, all of them or none of them: a scenario file on disk, or the scenario itself. Give exactly one of the two.
No parameter schema on file.
One call in full: headers and bodies, request and response. Header values that carry a session come back [REDACTED] unless raw is set.
No parameter schema on file.
Block until the client has made this call, instead of sleeping and hoping. Running out of time is not an error: the answer says timed_out and which seqs were seen.
No parameter schema on file.
Clear what one test case leaves for the next: script ctx.state and the replay counters, and with requests the request log as well. Neither flag means the state alone.
No parameter schema on file.
The switches a whole run is set by: how an endpoint with no rule is answered, which session recording and replay use, and what the request log keeps. Only the fields passed change.
No parameter schema on file.
Which hosts the proxy decrypts. Nothing HTTPS is read until its host is on the include list, so this comes before any rule for one. remove takes hosts off again, exclude edits the list that wins, enabled is the switch over both.
No parameter schema on file.
Keep the calls that reach the real server, to replay later. One switch over all passthrough traffic rather than a mode a rule is in: start it, stop it, or ask how it stands.
No parameter schema on file.
What has been recorded: one row per call, carrying the session and seq get_recording takes. This is what a rule in replay mode has to choose between.
No parameter schema on file.
One recorded call in full, headers and bodies. Read one before writing a script around ctx.recorded(): this is the shape that hands back.
No parameter schema on file.
Send one real request through the proxy port and show what came back, verified against tunnel's own CA — the way a device's request arrives. How a rule or a script is proved before anything is told to trust it.
No parameter schema on file.
No parameter schema on file.
- 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 tunnel or to call it, at any volume.
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.
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| 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 |
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.
Tools were statically extracted from the published source (16 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, 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. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| status | no tags | |
| guide | no tags | |
| list_mocks | no tags | |
| set_mock | no tags | |
| delete_mock | no tags | |
| apply_scenario | no tags | |
| list_requests | no tags | |
| get_request | no tags | |
| wait_for_request | no tags | |
| reset | no tags | |
| set_config | no tags | |
| decrypt | no tags | |
| record | no tags | |
| list_recordings | no tags | |
| get_recording | no tags | |
| check | no tags |
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
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 3.1.1. Pin a release below and the command asks for exactly that one.
No release note was published with this version.
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.
Any signed-in account that does not publish tunnel can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
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.