DogPolicies MCP Server
DogPolicies MCP server answers sourced, dated dog-travel questions for California and Nevada, and supports trip planning, highway-stop lookup, live Caltrans conditions, change monitoring, and correction reports.
Search dog-friendly places such as hotels, campgrounds, beaches, parks, trails, restaurants, and rest areas by name, type, county, region, area, state, leash rules, dogs-allowed status, or near a point.
Discover what dog-friendly places are near a town, neighborhood, region, or airport, grouped by kind.
Get full details for one place: rules, quotes, source URLs, confidence, freshness, verified dates, and facts the source does not state.
Compare two to six places side by side on dogs allowed, leash rules, fees, weight limits, pets per room, and restrictions.
Plan a whole dog trip: fitting stays, exact pet-fee arithmetic, day-by-day outings, optional dated plan and shareable trip link, plus an I-5 drive leg when relevant.
Find hotels that fit a dog’s weight, dog count, nights, and pet-fee budget, with match/possible/excluded results and fee math.
Look up chain-wide hotel pet policies for covered brands.
Browse ready-made answers to common dog-travel questions.
Find dog stops along highways in driving order, with mile, side, distance off the road, stay pet fees, and I-5 family/paired exits.
Explore I-5 by named stretch or mile window.
Check live I-5 conditions and live road conditions on I-80, US-50, I-15, and I-5 from Caltrans feeds.
Retrieve a specific rule by stable rule id with its quote, source, confidence, and verified date.
Monitor rule and verification changes since a date or near a place/route, and get Atom or JSON Feed links for future changes.
Report a correction as the only write operation, receiving a ticket id and public status page.
All answers include citations, verified dates, confidence/freshness metadata, and support markdown or structured JSON.
Uses OpenStreetMap to locate and count nearby places for people at freeway exits and pair them with the nearest dog-friendly DogPolicies stops; the OpenStreetMap-derived places are counted by kind and not named.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@DogPolicies MCP Server Can my 70 lb dog stay at a Monterey hotel for under $50 a night?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
DogPolicies MCP server
dogpolicies.com · Verified dog rules for California and Nevada travel: hotel pet fees, trip plans, highway stops. Install guide and example · Status
DogPolicies answers dog-travel questions for California and Nevada, and every fact it returns carries its official source (an agency page or the business's own published policy) and the date it was checked; Nevada coverage is still thin. It does the work a web page cannot: the pet-fee arithmetic for a specific dog, day-by-day trip plans, highway stops in driving order with Caltrans conditions on I-5 right now, what changed since a date, and corrections a person checks.
Listed in Claude's connector directory since November 4, 2026: add DogPolicies in Claude with nothing to install. Also on npm and in the official MCP Registry. DogPolicies is independent and not affiliated with Anthropic.
Connect
The hosted server is https://dogpolicies.com/mcp/: Streamable HTTP, no key, no account, nothing to install.
Claude (claude.ai or Claude Desktop): open DogPolicies in Claude's connector directory, add it and turn it on in the chat's tools menu. By hand: Settings → Connectors → Add custom connector, name DogPolicies, URL
https://dogpolicies.com/mcp/.Claude Code:
claude mcp add --transport http dogpolicies https://dogpolicies.com/mcp/Cursor and VS Code:
Both add the hosted URL; the app asks you to confirm.
Any other client that connects to a remote MCP server: give it the URL above.
To run it on your own machine over stdio instead (the data ships inside the package and works offline; Node.js 20.19 or newer):
{
"mcpServers": {
"dogpolicies": { "command": "npx", "args": ["-y", "dogpolicies-mcp"] }
}
}On Windows write "command": "cmd", "args": ["/c", "npx", "-y", "dogpolicies-mcp"], because clients start commands without a shell. Or let the installer write it:
npx dogpolicies-mcp install # Claude Desktop; --client cursor | vscode | windsurf | claude-code, or --all
npx dogpolicies-mcp install --hosted --dry-run # the URL form instead, previewed without changing anything
claude mcp add dogpolicies -- npx -y dogpolicies-mcp # Claude Code (Windows: -- cmd /c npx -y dogpolicies-mcp)The installer finds the client's config on Windows, macOS and Linux, backs it up, merges one entry without touching other settings, refuses a file it cannot parse and prints the restart step. Claude Desktop also takes the one-click extension, dogpolicies-mcp-<version>.mcpb, from the latest release. npx dogpolicies-mcp --http --port 3737 serves Streamable HTTP at http://127.0.0.1:3737/mcp.
Claude Desktop: Settings → Developer → Edit Config (
%APPDATA%\Claude\claude_desktop_config.jsonon Windows,~/Library/Application Support/Claude/claude_desktop_config.jsonon macOS), then fully quit and reopen it.Cursor:
~/.cursor/mcp.json, or.cursor/mcp.jsonin a project;{ "url": "https://dogpolicies.com/mcp/" }for the hosted server.VS Code:
~/.copilot/mcp-config.jsonwith the same block, orcode --add-mcp '{"name":"dogpolicies","type":"stdio","command":"npx","args":["-y","dogpolicies-mcp"]}'.Windsurf:
~/.codeium/windsurf/mcp_config.json;{ "serverUrl": "https://dogpolicies.com/mcp/" }for the hosted server.
Related MCP server: BeachFinder MCP
What it does
The tools, by the job they do. Every one except dogpolicies_report_correction only reads.
Fit and arithmetic
dogpolicies_find_hotels: which hotels take this dog, by weight, number of dogs, nights and a pet-fee budget. Each hotel is a match, a possible (call to confirm) or excluded, with the arithmetic spelled out ("2 dogs x 3 nights x $25 per pet per night = $150, capped at $100 per stay = $100"). Amounts are pet fees, never room rates.dogpolicies_plan_trip: a whole stay in one call. Stays that fit, with the pet fee for every night worked out, then the plan day by day: the stay, a morning beach, trail or off-leash area, an afternoon park and a dinner patio, each a real record with its source and date; a slot the data cannot fill is left empty with the reason.start_datedates each day,stay_idbuilds the plan on a chosen stay, andsave: truereturns ahttps://dogpolicies.com/trips/#v1.…link that carries the whole plan inside the link, so nothing is stored. Near I-5 it adds the drive leg.dogpolicies_get_brand_policy: a hotel chain's published pet policy (Best Western, Extended Stay America, Kimpton, La Quinta, Motel 6, Red Roof), its listed exceptions and how many properties carry it.
Routes and live state
dogpolicies_route_stops: dog stops along a highway in driving order, each with its mile, side and distance off the road: I-5, US-101, Highway 1, I-15, I-80 and US-50 in California; I-15, I-80, US-50, US-95 and US-93 in Nevada. Rest areas with pet areas, travel stops with dog parks, hotels with the stay's pet fee, campgrounds, parks and beaches. On I-5 also family stops (restrooms and a pet area stated by the source) and paired exits.dogpolicies_i5_corridor: I-5 in California by named stretch (the Grapevine, the Westside, Shasta and the rest), a mile range or the next miles ahead of a point.dogpolicies_i5_conditions: Caltrans lane closures, chain controls, road weather and sign messages on I-5 right now, each placed on the road beside the nearest dog stop. Live state, not rules; it needs the network.
A rest area Caltrans reports closed right now carries closed_now (Caltrans' reason and end time) in route_stops, i5_corridor and plan_trip, and is never offered as the next stop.
dogpolicies_road_conditions: Caltrans chain controls, lane and ramp closures, road weather and sign messages right now on I-80, US-50, I-15 and I-5 in California (Donner Summit, Echo Summit, Cajon Pass, the Grapevine), by route, town or named stretch, each at its route mile beside the nearest dog stop. Live state, not rules; Nevada roads are not covered; it needs the network.
Change
dogpolicies_changes_since: re-checks and rule changes since a date, newest first; on the hosted server also the corrections filed or decided since then, by ticket and status.dogpolicies_changes_near: the same around a town, region or point, for one place (place_id) or along a highway's stops (route,from_mile,to_mile), to re-check a trip before you go.dogpolicies_subscribe_changes: Atom and JSON Feed links that follow one place, a town or area, or a highway's stops for rule changes, with the last 90 days' counts and a sentence to hand to the traveler; it stores nothing and sends no email.
Comparison and citation
dogpolicies_near: everything dog-related near a town, neighborhood, region or airport, nearest first by kind; the usual first call. Names as people say them ("Van Nuys", "the Mission", "Lake Tahoe", "SFO", "LAX airport"); small spelling slips are corrected and labeled.dogpolicies_search_places: one named place, or a filtered list by type, county, region, area, dogs allowed or off leash. Free text such as "dog friendly hotels in Carmel" is read as the kind and the place.dogpolicies_browse_answers: ready-made answers to common questions ("Which California state parks allow dogs?").dogpolicies_get_place: one place in full: every fact with its source URL, publisher, quote and date checked, the rule ids, and what the source does not state.dogpolicies_compare_places: two to six places side by side (dogs allowed, leash, where dogs may go, pet fee, weight limit, pets per room), with a citation for each.dogpolicies_get_rule: one rule by its stable id, with the quote, source, confidence and verified date.
Action
dogpolicies_report_correction: files a correction for a person to check against the official source. The answer carries a ticket id and a public status page (https://dogpolicies.com/corrections/status/<ticket>/) that shows filed, then verified or rejected, with dates; what the traveler wrote is not shown. Nothing changes until it is verified. The only tool that writes.dogpolicies_plan_tripwithsave: true: a trip link to hand over (above).
Every answer
Starts with a citation sentence and ends with
DogPolicies, dogpolicies.com, CC BY 4.0, verified {date}. Every record carriesconfidence(stated, inferred_from_agency, inherited_brand),freshness(fresh, due, stale, unreadable),verified_at,urlandsource_urls. A fact the source does not state is listed as not stated, never filled in.Plain words work as arguments: "hotel", "NB", "Motel 6" or "yes" are read as the value the schema wants and labeled
resolved_from. An argument that still does not fit comes back asoutcome: "bad_input"with what is accepted, not a protocol error; a question the data cannot answer isno_matchwithdid_you_mean,nearestandnext.isErrormeans the server itself failed or refused a write (rate_limited).Markdown by default, structured JSON with
response_format: "json", and an output schema on every tool. The full reference, generated from the code: docs/tools.md.Resources attach a record without a tool call:
dogpolicies://place/{id},brand/{id},answer/{slug},route/{id}andarea/{id}(with completion), plusdogpolicies://about,answers,statusandcorridor/i-5. Prompts for the slash menu:plan_dog_trip,check_hotel_for_dog,road_trip_with_dog,drive_i5_with_dog,what_can_my_dog_do_here,cite_a_rule.
Not using MCP?
The read-only tools are also plain GET operations at https://dogpolicies.com/api/v1/ask/<operation>, run by the same engine with the same answers. The operation is the tool name without dogpolicies_, with hyphens (near, find-hotels, plan-trip, route-stops); arguments go in the query string, a point as near.lat=…&near.lon=…. The OpenAPI 3.1 description: https://dogpolicies.com/api/v1/openapi.json
curl "https://dogpolicies.com/api/v1/ask/find-hotels?county=Monterey&dog_weight_lb=70&max_total_fee=50"Status and data
dogpolicies.com/status shows the hosted server's version and commit, the data date, the last re-check, open corrections and the call tally; https://dogpolicies.com/api/v1/status.json is the same as JSON. The package ships a dated snapshot (as_of in every result), serves it at once, then checks https://dogpolicies.com/data/mcp-snapshot-v1.json in the background and swaps in a newer one when it exists. DOGPOLICIES_OFFLINE=1 (or --offline) never uses the network. All options: docs/configuration.md.
Try it in the terminal
npx dogpolicies-mcp demo # off-leash beaches near San Diego, with citations
npx dogpolicies-mcp demo "Monterey" # hotels in Monterey for a 70 lb dog, 2 nights, fees worked out
npx dogpolicies-mcp check # load the data and call every tool onceThe demo output is the exact text an assistant receives:
- Coronado Dog Beach (Coronado, CA): Yes, with restrictions. Dogs may be off leash only on the
designated Dog Beach area at 200 Ocean Boulevard and must be leashed at all points before
reaching it. Source: City of Coronado, checked 2026-10-26. DogPolicies record:
https://dogpolicies.com/places/coronado-dog-beach/ (CC BY 4.0; cite with link).
`coronado-dog-beach` · Coronado, San Diego County, California · 3 mi · freshness freshTroubleshooting
It does not show up. Restart properly: fully quit Claude Desktop (tray icon or Claude menu → Quit; closing the window leaves it running); in Cursor toggle the server in Settings → MCP; in VS Code run MCP: List Servers → Start; in Windsurf press Refresh in the MCP panel.
Check the server itself.
npx dogpolicies-mcp checkcalls every tool; if it works, the problem is the client's config or restart. Logs: Claude Desktop%APPDATA%\Claude\logs\mcp.logandmcp-server-dogpolicies.log(Windows) or~/Library/Logs/Claude/(macOS); Cursor: Output → MCP Logs; VS Code: MCP: List Servers → Show Output; Claude Code:claude mcp list."spawn npx ENOENT" on Windows. Use the
cmd /c npxform above (the installer writes it) or the hosted URL.Behind a proxy.
npxmust reach registry.npmjs.org once (setHTTPS_PROXY); the hosted URL and the extension need no npm.Still stuck? Open an issue with the output of
npx dogpolicies-mcp check, your OS and client: https://github.com/xultrax-web/dogpolicies-mcp/issues
Privacy
Nothing is collected. No telemetry, analytics or crash reporting; tool arguments are never logged above
debuglevel (stderr, your terminal).What leaves your machine with the local package: one conditional GET for the data snapshot on start and, only when you file a correction, the text you typed plus an optional contact you chose to give. Both go to dogpolicies.com over https and nowhere else;
--offlinestops both. The hosted server counts calls by tool and client name, never questions or addresses. Threat model and controls: docs/security.md.
License and attribution
Code: MIT.
Data: CC BY 4.0. Attribute "DogPolicies" with a link to https://dogpolicies.com/ and cite each record with its URL and verified date; every result includes a ready-made citation line. Chain-hotel names and locations come from OpenStreetMap under the ODbL and say so in
record_license.The rules are a sourced, dated summary, not legal advice. Confirm with the agency or the business before you go, especially when a record says
stale,unreadableorinherited_brand. Service animals are governed by the ADA, not these pet rules. Terms: https://dogpolicies.com/terms/
For developers
import { createRuntime, connectInProcess } from "dogpolicies-mcp";
const ctx = createRuntime({ overrides: { offline: true } });
const { client, close } = await connectInProcess(ctx);
const r = await client.callTool({
name: "dogpolicies_plan_trip",
arguments: { place: "Monterey", dog_weight_lb: 70, nights: 3 },
});
console.log(r.content[0].text);
await close();createServer(ctx) returns an McpServer from the official @modelcontextprotocol/server SDK for any transport. The reusable, dog-free core (dogpolicies-mcp/core) and how to build another server on it: docs/architecture.md. Versioning and deprecation: docs/versioning.md. Contributing: CONTRIBUTING.md.
Tests: npm test runs unit, contract, security, performance, surface, eval and fuzz tests; npm run test:e2e packs the tarball and runs it over stdio. CI runs on Linux, macOS and Windows with Node 20, 22 and 24.
Available Tools
17 toolsdogpolicies_browse_answersBrowse answered questionsARead-onlyIdempotentInspect
Which places match a common question, as a ready-made list: California national parks that allow dogs, off-leash dog beaches in California, state parks that allow dogs? Covers broad "list the ..." questions that are not about one town. Each list is fixed; it takes no county, type or off-leash filter. Without a slug it lists every question with its slug and result count; with a slug it returns that question's places (for example "off-leash-dog-beaches-california"), each with the dog rule, verified date and URL.
Inputs: slug, limit, offset, response_format. Returns: code, next, as_of, total, +11 more
dogpolicies_changes_nearWhat changed near a townARead-onlyIdempotentInspect
Has anything changed around where I am going since my last trip or since a date: a fee, a leash rule, a closure? Re-checks a saved itinerary or a destination: one area, one place or one stretch of highway. Inputs: place (city, county, neighborhood, area or region: "Tahoe", "Alameda", "Monterey") or near={lat, lon}, with radius_miles; or place_id for one place; or route (i-5, us-101, highway-1 ...) with from_mile and to_mile for the listed stops on that stretch. since as YYYY-MM-DD; optional outcome (confirmed, changed, source_gone, blocked, disputed). Returns verification events for places in scope, newest first, each with the place, outcome, date and record link, the rule-change log entries for them, and on the hosted server a separate corrections list (corrections filed for those places or decided since the date, by ticket id and status). An empty list means nothing was re-checked or changed in that window; an unknown name is an answer with did_you_mean.
Example: "What changed near Monterey since September?" -> place="Monterey", since="2026-09-01"
Inputs: since (required), near, limit, place, +8 more. Returns: code, next, as_of, count, +20 more
dogpolicies_changes_sinceChanges since a dateARead-onlyIdempotentInspect
What changed in the dog rules anywhere since a date: which fees, leash rules or closures were updated or found gone? A cheap poll for new changes across all the data, without re-reading records. Input: since as YYYY-MM-DD; optional outcome (confirmed, changed, source_gone, blocked, disputed), for example "changed" for only real rule changes. Returns events newest first, each naming the place, what happened, who checked, the date and a link to the record. On the hosted server a separate corrections list shows corrections people filed or whose status changed since the date: ticket id, place, kind of fact, status (filed, verified, rejected), dates and a status_url; what the reporter wrote is never included.
Inputs: since (required), limit, offset, outcome, +1 more. Returns: as_of, count, total, events, +8 more
dogpolicies_compare_placesCompare places side by sideARead-onlyIdempotentInspect
Which of these places is better for my dog: Carmel Beach or Asilomar, two parks, three hotels? Sets two to six known places side by side, fact by fact. Input: ids from search results ("carmel-beach", "asilomar-state-beach"); an exact name or a near miss resolves to the one clear match. Returns one row per fact (dogs allowed, leash rule, where dogs may go, pet fee, weight limit, pets per room, restrictions) and one column per place, with confidence, freshness, who confirmed it, and a citation for each. An id that matches nothing is listed under missing with close matches while the rest are still compared (outcome partial); fewer than two found is no_match with suggestions.
Inputs: ids (required), response_format. Returns: code, next, rows, as_of, +11 more
dogpolicies_find_hotelsFind hotels for a dogARead-onlyIdempotentInspect
Will a hotel take my dog, and what will the pet fee come to: a 70 lb Lab under $50, two dogs for three nights near Palm Springs, the Motel 6 in Fresno? Does the weight, dog-count and fee arithmetic for the stay, hotel by hotel, against each hotel's published pet policy. Inputs: query (hotel name, city or brand), brand, county without "County", state "CA" or "NV", area (lake-tahoe, big-sur, east-bay), region, or near={lat, lon}; dog_weight_lb, dogs, nights, max_total_fee. Each hotel is a match, a possible (needs a call: case by case, weight unstated, "up to" fee) or excluded, with the arithmetic spelled out, the source URL and verified date. Amounts are pet fees, never room rates: there are no room prices here.
Examples:
"Will a Monterey hotel take my 70 lb Lab for under $50?" -> county="Monterey", dog_weight_lb=70, max_total_fee=50
"Two dogs, three nights near Palm Springs" -> near={lat:33.83, lon:-116.55}, dogs=2, nights=3
"Is the Motel 6 in Fresno dog friendly?" -> query="Motel 6 Fresno"
"A hotel near LAX for my dog" -> query="LAX" (an airport code or name searches 5 miles around the airport)
"A Las Vegas hotel for my 60 lb dog" -> county="Clark", state="NV", dog_weight_lb=60
Inputs: area, dogs, near, brand, +11 more. Returns: code, next, as_of, count, +14 more
dogpolicies_get_brand_policyGet hotel brand pet policyARead-onlyIdempotentInspect
What is this hotel chain's pet policy: how many pets, weight limit, fee, exceptions? Answers chain-wide questions such as "What is Motel 6's pet policy?" or "Does Kimpton charge a pet fee?"; it does not work out one hotel's fee for a stay or say whether that hotel confirmed the chain policy. Brands covered: Best Western (best-western), Extended Stay America (extended-stay-america), Kimpton Hotels (kimpton), La Quinta by Wyndham (la-quinta), Motel 6 (motel-6), Red Roof Inn (red-roof). Input: brand id or name, e.g. "red-roof" or "Kimpton". Any other chain is not covered. Returns the published policy with the source URL, quote and verified date, and how many California and Nevada properties carry it.
Inputs: brand (required), response_format. Returns: id, url, code, name, +16 more
dogpolicies_get_placeGet place detailsARead-onlyIdempotentInspect
What are the exact dog rules for this one place, and what is the official source: the quote, the publisher, the date checked? The leash rule, where dogs may go, fees and limits in full, with the official source and its quote, for one place known by id or name. Input: id as returned in any result (for example "carmel-beach"); an exact name works, and a near miss resolves to the one clear match, labeled resolved_from. An id nothing matches is an answer with did_you_mean, not an error. Returns every rule with display text and rule ids, each source with URL, publisher, capture date and quote, and facts the source does not state listed as not_stated. A place on I-5 in California also carries its position on the road (Caltrans mile, side, distance off the freeway) and any paired exit.
Inputs: id (required), response_format. Returns: id, uid, url, axis, +32 more
dogpolicies_get_ruleGet one rule by idARead-onlyIdempotentInspect
Where does this specific rule come from (a fee, a leash rule, a weight limit): what is the exact quote, the source and the date it was verified? Takes one rule id and returns the evidence behind that single fact. Rule ids look like "rul__" and appear in every place result as rule_records. Returns the rule's value, the source it cites with the quote, confidence, verified date, freshness and the place's citation line.
Inputs: id (required), response_format. Returns: code, next, rule, as_of, +10 more
dogpolicies_i5_conditionsI-5 conditions nowARead-onlyIdempotentInspect
Is I-5 open right now: lane closures, chain controls, road weather and sign messages on the freeway in California, and the nearest dog stop to each? Live state, not rules: Caltrans feeds fetched within the last 15 minutes (fetched_at in the result), not verified rules. Covers the Grapevine and every other stretch of I-5 in California. Other highways are not covered. Inputs: from_mile and to_mile, or near={lat, lon} with direction and ahead_miles; with none, all of I-5 in California. Each condition is placed on the I-5 centerline (mile, s meters from San Ysidro) and matched to the nearest dog stop. If the feeds cannot be read the answer has outcome unavailable and the address of the live Caltrans page. A result older than two hours is marked stale.
Examples:
"Is there a closure or chains on the Grapevine right now?" -> from_mile=170, to_mile=222
"Weather and closures in the next 60 miles northbound from Redding" -> near={lat:40.59, lon:-122.39}, direction="north", ahead_miles=60
Inputs: near, to_mile, direction, from_mile, +2 more. Returns: code, next, as_of, signs, +19 more
dogpolicies_i5_corridorI-5 corridor by stretchARead-onlyIdempotentInspect
What is on I-5 between here and there for a dog: the next restroom stops with a pet area, the stops on the Grapevine, between Sacramento and Redding? For the freeway in California only, by one of eleven named stretches (sandiego, orange-la, grapevine, kern-westside, westside, delta, sacramento, valley, tehama, shasta, siskiyou), a mile range, or a window ahead of a point. It covers I-5 only, with no hotel fee arithmetic and no closures or chain controls. Inputs: strip, or from_mile and to_mile (Caltrans odometer), or near={lat, lon} with direction (north or south) and ahead_miles; family=true keeps stops whose official source states restrooms and a pet area or dog park; paired=true adds exits with places for people nearby (counted from OpenStreetMap, never named, not verified for restrooms) matched to the nearest dog-legal place. With none of these it lists the stretches. Each stop carries its mile, s in meters from San Ysidro, the side of the road, distance off the freeway, and restrooms, water, pet area and hours as the source states them, with source URL and verified date. A rest area Caltrans reports fully closed right now carries closed_now (Caltrans' reason and end time); it stays in the list and the text names the first open stop instead.
Examples:
"What is between the Grapevine and Kettleman?" -> strip="grapevine" and strip="kern-westside"
"The next 5 family stops northbound from Stockton" -> near={lat:37.95, lon:-121.29}, direction="north", family=true, limit=5
"Every stop between mile 300 and 340" -> from_mile=300, to_mile=340
Inputs: near, kinds, limit, strip, +9 more. Returns: code, next, as_of, stops, +15 more
dogpolicies_nearWhat is near a townARead-onlyIdempotentInspect
What can my dog do in or around this town, neighborhood or region: beaches, parks, trails, hotels, campgrounds, cabins, patios and rest areas, nearest first? Answers "dog-friendly things in X" questions, grouped by kind. It does not check hotels against a dog's weight or a fee budget. Input: place as people say it: a city, county, neighborhood or alias ("Van Nuys", "Koreatown", "Venice Beach", "the Mission", "DTLA"), an area or region ("Lake Tahoe", "East Bay", "Central Valley"), or near={lat, lon}. Small spelling slips ("Montery") are corrected and labeled. If nothing is inside the radius it widens to 30, 60, then 100 miles and says so. Each result carries the dog rule, verified date and source URL. An unknown name comes back as an answer with did_you_mean, the nearest towns and next (a larger town or county that is covered).
Examples:
"What can my dog do in Carmel?" -> place="Carmel"
"Dog beaches and parks around South Lake Tahoe" -> place="South Lake Tahoe", types=["beach","park"]
"Anything dog-friendly in Van Nuys?" -> place="Van Nuys"
"Anything dog-friendly within 5 miles of here" -> near={lat:36.6, lon:-121.9}, radius_miles=5
Inputs: near, place, types, per_type, +3 more. Returns: code, next, as_of, label, +14 more
dogpolicies_plan_tripPlan a trip with a dogARead-onlyIdempotentInspect
Plan a dog trip to a town: where can we stay with this dog, what is the pet fee for all the nights, what can we do each day, where can we eat? One answer covers a whole stay, such as "three nights in Monterey with our 70 lb Lab, pet fees under $150" or "camping near Mammoth Lakes". Inputs: place (a city, county, neighborhood, area or region, with small spelling slips corrected) or near={lat, lon}; dog_weight_lb, dogs, nights, max_total_fee, stay_types. Optional: start_date (dates each day), stay_id (build the plan on that stay), save=true (adds trip_url, a dogpolicies.com page showing the plan; the plan travels inside the link, nothing is stored). Returns stays that fit with the whole-stay pet fee worked out (a pet fee, never a room rate), stays that need a call to confirm, and "days": for each night the stay, a morning beach, trail or off-leash area, an afternoon park and a dinner patio, each a real record with source URL and verified date; a slot the data cannot fill is null with the reason (dinner often is). "plan_stay" shows the fee arithmetic and whether the dog's weight and number fit. Within 10 miles of I-5 in California it adds the drive leg: the town's mile and the next family stops each way. An unknown town is an answer with did_you_mean. The markdown answer is laid out as a trip document for the traveler.
Examples:
"Three nights in Monterey with our 70 lb Lab, pet fees under $100" -> place="Monterey", dog_weight_lb=70, nights=3, max_total_fee=100
"A weekend in South Lake Tahoe with two small dogs" -> place="South Lake Tahoe", dogs=2, nights=2
"Camping near Mammoth Lakes with a dog" -> place="Mono County", stay_types=["campground","cabins"]
"Plan 4 nights in Carmel with two dogs and give me a link" -> place="Carmel-by-the-Sea", dogs=2, nights=4, save=true
Inputs: dogs, near, save, place, +10 more. Returns: i5, code, days, next, +19 more
dogpolicies_report_correctionReport a correctionAInspect
Tell us a rule is wrong: the fee went up, a beach closed to dogs, a hotel now takes two pets. For a record the traveler says is wrong or out of date; this is the one tool that writes. The report goes to a person who re-reads the official source, and the record does not change until it is verified: a report is filed, not a fix. Inputs: place_id from a result, message (what is wrong, in the traveler's words), optional fact (which kind of fact is wrong), optional source_url of the official page, optional contact if the traveler offers one. No other personal detail is accepted. The answer carries a ticket id and a status_url. The ticket's status (filed, then verified or rejected) is public there; what they wrote is not.
Inputs: message (required), place_id (required), fact, contact, +1 more. Returns: url, code, next, status, +11 more
dogpolicies_road_conditionsRoad conditions now: closures, chains, weatherARead-onlyIdempotentInspect
Do I need chains on I-80 to Tahoe right now? Is US-50 open over Echo Summit? Caltrans chain controls, lane and ramp closures, road-weather stations and message-sign text on I-80 (San Francisco to the Nevada line), US-50 (West Sacramento to Stateline), I-15 (San Diego to Primm) and I-5 in California, each at its route mile with the nearest dog stop. Live state, not rules: Caltrans feed data fetched every 15 minutes (fetched_at in the result), never a verified fact. For the day of a mountain or desert drive: Donner Summit, Echo Summit, Cajon Pass, the Grapevine. It reports conditions, not places to stop: each condition carries only the nearest dog stop. Nevada roads and other highways are not covered. Inputs: route ("I-80", "80", "US 50", "Highway 50", "I-15", "I-5"); a window as from_mile and to_mile, or town (a town or named stretch on the route: Truckee, Donner Pass, South Lake Tahoe, Placerville, Cajon Pass, Barstow) with direction and ahead_miles; direction alone keeps that direction's closures, signs and chain controls. With no window, the whole route in California. chain_controls_in_effect is true when a Caltrans chain-control point in the window reports a control (any status but R-0), with Caltrans' own wording; a sign that mentions chains is listed under signs with its text. Each weather station says whether its reading is current (the feeds sometimes carry readings months old; those numbers are left out of the text). A document older than two hours is marked stale; when no document can be read the answer has outcome unavailable and the address of Caltrans QuickMap, the live map.
Examples:
"Do I need chains on I-80 to Tahoe right now?" -> route="I-80", direction="east"
"Is Highway 50 open over Echo Summit?" -> route="Highway 50", town="Echo Summit"
"Closures on I-15 from Barstow to the Nevada line" -> route="I-15", town="Barstow", direction="north", ahead_miles=120
Inputs: route (required), town, to_mile, direction, +3 more. Returns: code, next, road, as_of, +24 more
dogpolicies_route_stopsDog stops along a highwayARead-onlyIdempotentInspect
Where can we stop with the dog along this highway, in driving order: rest areas, travel stops with dog parks, hotels, campgrounds, parks and beaches, each with its mile and distance off the road? For a drive on I-5, US-101, Highway 1, I-15, I-80 or US-50 in California, or I-15, I-80, US-50, US-95 or US-93 in Nevada (each route stops at the state line). It lists stops, not closures or chain controls. Inputs: route (default i-5), a start as near={lat, lon} or from_mile, direction (north, south, east, west), ahead_miles or to_mile, dog_weight_lb and max_total_fee for stays. On I-5 add family=true (restrooms and a pet area stated by the source) or paired=true (an exit with places for people, plus the nearest dog-legal place). Each stop has its mile, side, distance off the road, dog rule, source URL and verified date; stays carry the whole-stay pet fee. On I-5 a rest area Caltrans reports fully closed right now carries closed_now (Caltrans' reason and end time), is left out of the "first open stop" in the text, and stays in the list.
Examples:
"Driving north on I-5 from Los Angeles with a 70 lb dog, where can we stop?" -> near={lat:34.05, lon:-118.24}, direction="north", dog_weight_lb=70
"Pet areas on the Grapevine" -> from_mile=170, to_mile=222, kinds=["rest_area","park"]
"Next restroom stop for the whole family, dog included, 40 miles north of Harris Ranch" -> near={lat:36.25, lon:-120.25}, direction="north", ahead_miles=40, family=true
"Heading south on 101 from San Francisco, where can the dog run?" -> route="us-101", near={lat:37.77, lon:-122.42}, direction="south"
"Driving I-80 east from Reno with a dog" -> route="i-80-nevada", near={lat:39.53, lon:-119.81}, direction="east"
Inputs: dogs, near, kinds, limit, +13 more. Returns: code, next, as_of, count, +17 more
dogpolicies_search_placesSearch placesARead-onlyIdempotentInspect
Can my dog go to this place, or which parks, beaches, trails, campgrounds or patios allow dogs: "Can I bring my dog to Point Lobos?", "state parks in Marin that allow dogs", "off-leash beaches near San Diego"? Answers for one named place or a filtered list of one kind; each result's id names the place's full record, with every rule and source. Inputs: query (name, city or brand; words like "cabins", "hikes", "wineries" or "dog friendly hotels in Carmel" are read as the kind and the place), type (lodging, campground incl. RV parks, cabins, beach, park, trail, restaurant, rest_area, stop; "hotel", "dog park" or "hike" are read as these), dogs_allowed, off_leash, state "CA" or "NV", county without "County", area (lake-tahoe, big-sur, east-bay), or near={lat, lon}. Each result: yes, yes_restricted, no or varies, the rule in a sentence, verified date, URL and id. No match is an answer (outcome no_match) with did_you_mean, nearest and next. Nevada coverage is thin.
Examples:
"Off-leash beaches near San Diego" -> type="beach", off_leash=true, near={lat:32.72, lon:-117.16}
"Can I bring my dog to Point Lobos?" -> query="Point Lobos"
"State parks in Marin County that allow dogs" -> type="park", county="Marin", dogs_allowed=true
"Lake Tahoe beaches in Nevada that allow dogs" -> type="beach", state="NV", dogs_allowed=true
"Off-leash dog parks in Las Vegas" -> type="park", off_leash=true, county="Clark", state="NV"
"RV parks in Fresno County that take dogs" -> type="campground", county="Fresno", dogs_allowed=true
"Dog-friendly cabins in June Lake" -> type="cabins", query="June Lake", dogs_allowed=true (query="June Lake cabins" works too)
"Restaurants with a dog-friendly patio in Alameda" -> type="restaurant", query="Alameda"
Inputs: area, near, type, limit, +8 more. Returns: code, next, as_of, count, +13 more
dogpolicies_subscribe_changesFollow rule changes in a feedARead-onlyIdempotentInspect
Can you tell me if the dog rules change at Carmel Beach before our trip? Can I follow changes near Lake Tahoe, or along I-5? For hearing about changes later, without asking again: it answers with a feed link (Atom and JSON Feed) for one place, a town, county, area or neighborhood with a radius, or a highway's stops, which any feed reader can follow. It computes the links and counts only; it signs nobody up, stores nothing and sends no email. Inputs: one of place (a place id from any answer, e.g. "carmel-beach"), near (a town, county, area, region or neighborhood, e.g. "Lake Tahoe", "Monterey", "Van Nuys") with radius_miles, or route (i-5, us-101, highway-1, i-15, i-80, us-50, or a Nevada route ending in -nevada) with from_mile and to_mile. Returns atom_url and json_url, how many verification events, rule changes and corrections the scope had in the last 90 days, the page for the scope, and share_text: a sentence with the link, written for the traveler. An unknown name is an answer with did_you_mean.
Examples:
"Tell me if the rules change at Carmel Beach" -> place="carmel-beach"
"Can I get alerts for dog rule changes around Lake Tahoe?" -> near="Lake Tahoe"
"Follow changes on the Grapevine stretch of I-5" -> route="i-5", from_mile=170, to_mile=222
Inputs: near, place, route, to_mile, +3 more. Returns: code, next, as_of, scope, +14 more
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 17 tools
Most tools have clearly distinct purposes, and the descriptions explicitly draw boundaries between search, detail, comparison, planning, and evidence lookup. There is some overlap among near/search_places/browse_answers and especially route_stops vs i5_corridor and i5_conditions vs road_conditions, so an agent may occasionally need to read carefully before choosing.
All tools use a consistent dogpolicies_ prefix and snake_case, which makes the set predictable. However, naming mixes verb-first patterns like search_places, get_place, and plan_trip with noun-only or noun-phrase names like near, route_stops, i5_conditions, and road_conditions.
17 tools is slightly above the typical ideal range, but the domain is broad enough to justify most of them: discovery, rules, hotels, trip planning, highway conditions, changes, feeds, and corrections. A few overlapping highway and change-monitoring tools could potentially be consolidated, but the count is not unreasonable.
The surface covers the full dog-travel lifecycle: discovery, exact rule evidence, comparison, trip planning, hotel and brand policies, curated lists, highway stops, live road conditions, change monitoring, subscriptions, and correction reporting. There are no obvious dead ends for the stated domain.
Maintenance
Related MCP Connectors
U.S. law library for AI: court decisions, statutes, court rules and citation checks. Read-only.
Staylist World is a hotel comparison service your AI assistant can call directly. The free MCP server (https://staylist.world/mcp, Streamable HTTP, no auth) covers 80,000+ hotels with ratings, review summaries, published policies and booking links. No API key, no sign-up. Tools: search_hotels by city, name or need; get_hotel for overviews, amenities and published policies; get_booking_link; list_destinations; and get_reviews. There is also a REST API and Markdown versions of hotel pages. Unknown facts are marked as unknown. New: get_reviews returns a hotel's 1–2 star guest reviews from the last 24 months with the count out of all reviews; or paste staylist.world/reviews?hotel=<name> into any assistant.
Dated, sourced AI API prices, model retirements and AI app facts. Read-only, no key needed.
Give AI assistants access to real-time data. Search the web, compare flights, find hotels, and more.
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables AI assistants to search for pet sitters, manage bookings, and handle pet profiles through the Rover pet services marketplace. It facilitates direct communication with sitters and service management using Playwright-based browser automation.132MIT
- FlicenseNot gradedqualityDmaintenanceProvides AI assistants read-only access to discover and compare 184,900+ beaches, lakes, and swimming spots worldwide with current planning signals.-
- FlicenseNot gradedqualityDmaintenanceEnables AI assistants to fetch dog breed data, including descriptions, attributes, and group information, via the Dog API through natural language queries.-
- AlicenseNot gradedqualityCmaintenanceEnables users to search dog breeds, retrieve detailed breed information, get random dog facts, and access AKC breed groups through the dogapi.dog v2 API.88 npmMIT