Comprehensive Model Context Protocol server for managing Home Assistant through AI assistants.
Nothing to run from here
gmaps.dev 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.
93/100
tool-list score
18
tools declared
remote
at its publisher
—
calls last 7d
What it does
Find businesses by what they are and where they are, collect a whole city, and export the list.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --transport http gmaps-dev-mcp https://api.gmaps.dev/mcp
# 2 — ask your agent for something these tools do
# get_attributions, search_places, get_place and 15 more
gmaps.dev costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
GMAll 1 server
gmaps1 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. Nothing here runs on your machine and nothing runs on ours: this server answers at its publisher’s own address, and what you install is the connection straight to it.
Claude Codedeclared by this listing
This is a public server: it answers at its publisher’s own address and this marketplace is not in the path. Claude Code registers it in one command; --scope user makes it available in every project. The transport below is http, as the publisher's own manifest declares it.
claude mcp add --scope user --transport http gmaps-dev-mcp https://api.gmaps.dev/mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it answers at its publisher’s own address: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it writes
One config entry pointing straight at the publisher’s own address — this marketplace is not in the path, and nothing from this listing executes on your machine.
What you approve
18 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.
List the projects credited by gmaps.dev, with their licenses and notices. No API key needed. Costs no credits.
No parameter schema on file.
Find places by what they are and where they are, such as bakeries in Curitiba. Places we saved from earlier searches come back at once, for free. If they do not fill limit, a collection starts to find the rest and the answer includes it: each new place costs one credit as it arrives, and it stops at limit, so a search never costs more than limit credits. Search again later, or read the colle
| Parameter | Type | Required | Notes |
|---|---|---|---|
| keywords | array | yes | What to search for, such as ["bakery"]. One search per term. |
| geo | yes | Where to search. Give exactly one of: a Brazilian city as location, a lat/lng point with radiusM, a bbox, or a cityIbge with radiusKm. | |
| lang | string | no | The language to search in. New places come back in it. Two letters, such as pt or en. |
| minRating | no | Keep only places rated at least this, from 1 to 5. Null keeps every place. | |
| limit | integer | no | How many places you want, from 1 to 120. Also the most a search can cost. |
| wait | integer | no | Seconds to wait for new places when we have to collect them, up to 60. 0 answers at once with what we already have. |
Read one saved place, by our id, Google's cid, Google's place ID, or a Google Maps link to it. It never searches: a place we have not saved answers 404, and searching its area with search_places brings it in. Needs an API key. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| place | string | yes | Our id for the place, Google's cid, Google's place ID, or a Google Maps link to it. |
Search a whole area, square by square, and collect up to maxPlaces of the places in it. Answers straight away with the collection, then works through the area in the background. Read the collection again to see how far it has got. Starting one costs nothing: each new place costs one credit as it arrives, and saved places cost nothing. Needs an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| keywords | array | yes | What to search for, such as ["bakery", "coffee shop"]. One search per term. |
| geo | yes | Where to search. Give exactly one of: a Brazilian city as location, a lat/lng point with radiusM, a bbox, or a cityIbge with radiusKm. | |
| lang | string | no | The language to search in. New places come back in it. Two letters, such as pt or en. |
| minRating | no | Keep only places rated at least this, from 1 to 5. Null keeps every place. | |
| maxPlaces | integer | no | Stop after this many places. Defaults to the most your plan collects at once. |
| enrich | array | no | Add ["emails"] to also look for email addresses. Costs one more credit for each place that has one, and needs a paid plan. |
Your collections, newest first, with the total so you know how many more there are. Filter by status to find the ones still running. Needs an API key. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| status | string | no | Show only collections with this status, such as running. |
| limit | integer | no | How many collections to return, from 1 to 100. |
| offset | integer | no | How many collections to skip, for reading the next page. |
See where a collection has got to: how much of the area is searched, how many places have arrived, and what it has cost so far. Read it again whenever you want an update. Needs an API key. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | The collection's id, as create_collection returned it. |
Stop a collection you no longer want. One that has not started yet stops at once and costs nothing. One already running keeps the places that have arrived and the credits they cost; the search in flight finishes, and nothing new starts. A collection that has already ended cannot be cancelled. Needs an API key. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | The collection's id, as create_collection returned it. |
See what collecting an area would cost before you start it: how much of it we have searched recently, how many places we have saved there (those cost nothing), how many searches are left to run, and the most it can cost in credits. Needs an API key. Costs no credits and starts nothing.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| keywords | array | yes | What to search for, such as ["bakery", "coffee shop"]. One search per term. |
| geo | yes | Where to search. Give exactly one of: a Brazilian city as location, a lat/lng point with radiusM, a bbox, or a cityIbge with radiusKm. | |
| lang | string | no | The language to search in. New places come back in it. Two letters, such as pt or en. |
| minRating | no | Keep only places rated at least this, from 1 to 5. Null keeps every place. | |
| maxPlaces | integer | no | Stop after this many places. Defaults to the most your plan collects at once. |
| enrich | array | no | Add ["emails"] to also look for email addresses. Costs one more credit for each place that has one, and needs a paid plan. |
Get a link to a collection's places as one file: csv, json or xlsx. The file has the places the collection holds now, even if it is still running. The link opens without an API key for 24 hours, so share it only with people who should see the file. Needs an API key. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| collectionId | string | yes | The collection to export, as create_collection or search_places returned it. |
| format | string | no | The file type: csv, json or xlsx. An xlsx file holds up to 50,000 places. |
| locale | string | no | The language of the column names in csv and xlsx files: en or pt-BR. Defaults to your account's language. |
Read an export you made, with its download link, while the link still works. After 24 hours, or once the collection's places are deleted, it answers 404: make a new export from the collection. Needs an API key. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | The export's id, as create_export returned it. |
See how much of an area we have already searched for a term, when we last searched it, and how many places we have saved there. Use it to decide whether to collect again or just read the saved places. Needs an API key. Costs no credits and starts nothing.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| keywords | array | yes | What you would search for, such as ["bakery"]. Coverage is per term: we may have searched an area for one and never for another. |
| geo | yes | The area. Give exactly one of: a Brazilian city as location, a lat/lng point with radiusM, a bbox, or a cityIbge with radiusKm. | |
| lang | string | no | The language the search would run in. Two letters, such as pt or en. |
See this month's credits: what your plan includes, what you have used, what is left today and this month, and the same totals for each API key. Needs an API key. Costs no credits.
No parameter schema on file.
Tell us where to POST when a collection ends, instead of asking us every few seconds whether it is done. The answer shows the signing secret once and never again. Store it, then check every request against it. The body carries counts and a link to the places, never the places themselves. Needs an API key and a paid plan. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| url | string | yes | Where to POST the event. Must start with https:// and resolve to a public address. We will not send your results unencrypted or into a private network. |
| events | array | no | Which events to send here. "collection.completed" fires when a collection ends, even one that found no places; "collection.failed" fires when an error stops it. A collection you cancel sends neither. |
| description | string | no | Your own note about this endpoint, such as which of your services it is. |
Every endpoint you have added, newest first, with whether it is still receiving events and why we stopped if it is not. Secrets are not shown again after the endpoint is created. Needs an API key and a paid plan. Costs no credits.
No parameter schema on file.
Change which events an endpoint receives, rename it, or switch it off and on. This is how an endpoint comes back after we stopped sending to it: fix your server, then set enabled to true. The failure count starts again from zero. Whatever you leave out stays as it was. Needs an API key and a paid plan. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | The endpoint's id, as create_webhook returned it. |
| enabled | boolean | no | False to stop sending here without deleting it. True to start again after we switched it off. |
| events | array | no | Which events to send here. "collection.completed" fires when a collection ends, even one that found no places; "collection.failed" fires when an error stops it. A collection you cancel sends neither. |
| description | no | — |
Stop sending to an endpoint and forget it, along with the record of what we sent there. To pause one instead, set enabled to false. That keeps the secret, so you do not have to change your code to turn it back on. Needs an API key and a paid plan. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | The endpoint's id, as create_webhook returned it. |
Send a "ping" to one of your endpoints now, so you can check your code reads it and the signature checks out before a real collection depends on it. Answers straight away with the delivery; read the deliveries list a moment later to see what your server said. Needs an API key and a paid plan. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | The endpoint's id, as create_webhook returned it. |
| message | string | no | What the test event should say, so you can spot it in your logs. |
What we sent to one endpoint, newest first: which event, whether it arrived, how many tries it took, and what your server answered. This is where to look when an event never showed up. Kept for 7 days. Needs an API key and a paid plan. Costs no credits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | The endpoint's id, as create_webhook returned it. |
| limit | integer | no | How many deliveries to return, newest first, from 1 to 100. |
Free
$0 forever
- Every tool — any call limit is its publisher’s, not ours
- No card, and mcprush charges nothing for it
- Runs at its publisher’s own address — this marketplace is not in the path
- 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 gmaps.dev 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: —
ATool-list grade · score 93 / 100
Release 1.0.0 · scanned 22 Oct 2026
Graded from the tools this server lists, not from its source code.
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 score93
Findings0 threat · 2 capability
Capabilitymoderate
Coveragemanifest
Verificationnot in this report
Surface digest41ad28d6050ef033fa1c6261a3c33ee4d9dc45a2ab2ffda58b12cdea55318cd1
Score vectorthreat 100 → client 93
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | moderate | capability blast radius (moderate) — client exposure if the model is manipulated | −3 |
| coverage-honesty | manifest | inspection depth (manifest) — how much of the target the scan could see | −4 |
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.
The scan read each tool’s name, description, input schema and annotations exactly as the server listed them the last time it answered us; no code was reviewed. We ask the server again about once a month, and a list that changed is graded again.
Tool surfaceread
Implementation sourcenot available
Package metadatanot available
Live transportnot available
The implementation source was not read — this reflects the declared tool metadata only.
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. moderate is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| get_attributions | no tags | |
| search_places | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| get_place | no tags | |
| create_collection | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| list_collections | no tags | |
| get_collection | no tags | |
| cancel_collection | state-write | state-write: declares readOnlyHint: false, destructiveHint: true |
| quote_collection | no tags | |
| create_export | state-write | state-write: declares readOnlyHint: false |
| get_export | untrusted-input | untrusted-input: keyword "download" |
| get_coverage | no tags | |
| get_usage | no tags | |
| create_webhook | external-sink | external-sink: keyword "webhook", declares readOnlyHint: false, openWorldHint: true |
| list_webhooks | no tags | |
| update_webhook | external-sink | external-sink: keyword "webhook", declares readOnlyHint: false, destructiveHint: true, openWorldHint: true |
| delete_webhook | external-sinkstate-write | external-sink: keyword "webhook"; state-write: declares readOnlyHint: false, destructiveHint: true |
| test_webhook | external-sink | external-sink: keyword "webhook", declares readOnlyHint: false, destructiveHint: true, openWorldHint: true |
| list_webhook_deliveries | 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.
flow-1
Untrusted input can drive an external action even though no sensitive source is exposed.
Supply chain and provenance
Integrityfirst-seen
Surface digest41ad28d6050ef033
Scanned2026-10-22
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
This is an address its publisher runs: which release it serves is theirs to choose, so there is nothing here to pin. The history below is for reading; the current release on file is 1.0.0.
1.0.0
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 gmaps.dev 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.