Solano — marine & outdoor weather MCP server
fr.solanometeo/solanosolanometeo.frstreamable-httpv1.2.0Scanned
Marine and outdoor weather: multi-model forecasts, official bulletins, tides, anchorages.
9 tools: 9 only read.
- Reads private data · 0
- Reads text others write · 9
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
Decide which of this server's tools your agent can use.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Solano — marine & outdoor weather MCP server can do
The Solano — marine & outdoor weather MCP server is published by solanometeo.fr, a domain the MCP Registry verified. It has 9 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
9 tools bring in text other people can write (get_forecast, get_confidence, get_storm_risk and 6 more).
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in, and its tools changed 2 times since Anlyon Radar first saw them.
Last scanned Nov 15, 2026, version 1.2.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
9 tools
Only reads 9Show 9 read-only tools
analyze_anchorageDirectional shelter analysis of ONE precise anchorage: how protected it is from every compass direction, and how comfortable it will be over the next three days.Reads text others writeNo findings
Why it is in this group
- The name "analyze_anchorage" says "analyze": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote itContains invisible characters, shown as [U+…] Directional shelter analysis of ONE precise anchorage: how protected it is from every compass direction, and how comfortable it will be over the next three days. Casts 1728 rays against a global elevation model plus OSM breakwaters and moles, giving per-direction wind protection and whether swell is blocked; then scores comfort hour by hour against the real forecast. Also returns depth (EMODnet bathymetry) and seabed nature: a seagrass meadow (`seabed.seagrass`, e.g. protected Posidonia) always takes precedence over the sediment beneath (`seabed.sediment`) and means anchoring there should be avoided. `anchoring_zones` lists the mapped areas containing the spot where anchoring is prohibited or restricted (OpenStreetMap, volunteer-mapped): pass them on. An empty list does NOT mean anchoring is allowed — coverage is patchy. ⚠[U+FE0F] Use this on a SINGLE spot the user has in mind. It is the most expensive thing this server does — 18 topography calls against a public quota shared with the website — so it is capped per day, serialised, and de-duplicated. Do NOT call it repeatedly to sweep a coastline: call `find_known_anchorages` for that, or send the user to the app, which has a search built for it. A refusal here is a budget limit, not an error. Points already analysed come back from cache instantly and cost nothing.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
find_known_anchoragesAnchorages already analysed nearby, re-scored for comfort over the next three nights against the real forecast wind and swell.Reads text others writeNo findings
Why it is in this group
- The name "find_known_anchorages" says "find": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Anchorages already analysed nearby, re-scored for comfort over the next three nights against the real forecast wind and swell. Only spots whose full shelter profile is already in Solano's shared cache are considered — work other users' analyses have already paid for. Each is scored per night (18:00 to 10:00 local) on the worst hour of the window, which is the conservative reading you want for a night at anchor: swell weighs 0.70, wind 0.30, because swell is what makes a boat roll. Empty result means nobody has analysed an anchorage around this point yet, not that there is none — use `analyze_anchorage` on a precise spot to add one, or open the app to sweep the area. Cheap by construction: no topography download, two forecast calls.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
get_confidenceHow much the forecast can be trusted at this point, hour by hour, from ensemble spread — Solano's own indicator, not a raw model output.Reads text others writeNo findings
Why it is in this group
- The name "get_confidence" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it How much the forecast can be trusted at this point, hour by hour, from ensemble spread — Solano's own indicator, not a raw model output. Each deterministic model is compared against the members of ITS OWN reference ensemble (GFS→GEFS, ECMWF IFS→ECMWF-ENS, ICON→ICON-EPS), never against a mixed pool: the systematic offset between forecasting centres would otherwise be charged to the model as if it were uncertainty. The percentage blends wind (60%), rain intensity class (25%) and pressure (15%). High means the scenarios converge, so the forecast is solid; low means they diverge and the forecast should be treated as uncertain. Use this when asked "is this forecast reliable?", "will it hold?", or before committing to a passage plan. Returned per day: mean and minimum confidence, plus the per-variable breakdown at the least confident hour.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
get_forecastDaily weather summary for a point, from one or more international models.Reads text others writeNo findings
Why it is in this group
- The name "get_forecast" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Daily weather summary for a point, from one or more international models. Returns, per model and per LOCAL day: mean and max wind (knots), max gust, dominant wind direction, min/max temperature, total precipitation, and peak CAPE. Not hour-by-hour — follow `solano_url` for the hourly view. When the point is at sea and `include_sea_state` is true, a `sea_state` block adds significant wave height (Hs), derived maximum wave height (Hmax), period and direction from a wave model. Models are picked for the location: a regional high-resolution model (AROME 1.5 km, ICON-2I, ICON-D2, UKV…) replaces its global parent where it is available. Pass `models` to force specific keys — use `list_models_for_location` first to see what exists there.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
get_marine_bulletinThe official marine forecast bulletin for the maritime zone containing a point — the text a national weather service publishes for mariners, not a model output.Reads text others writeNo findings
Why it is in this group
- The name "get_marine_bulletin" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it The official marine forecast bulletin for the maritime zone containing a point — the text a national weather service publishes for mariners, not a model output. Providers, matched automatically to the point: Météo-France (BMR coastal), AEMET (Spain), MeteoAM (Italy), Met Office (UK inshore waters and shipping forecast). Pass `provider` to force one of `mf`, `aemet`, `meteoam`, `metoffice`. The bulletin is returned in the language its issuing service publishes it in (see `language`). Quote it, do not translate it as if it were the official wording. Costs no third-party call when warm: bulletins are cached 1-3 hours.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
get_storm_riskThunderstorm environment at a point, scored 0-10 per hour from convective ingredients rather than from a rain field.Reads text others writeNo findings
Why it is in this group
- The name "get_storm_risk" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote itContains invisible characters, shown as [U+…] Thunderstorm environment at a point, scored 0-10 per hour from convective ingredients rather than from a rain field. Built on a GFS baseline that carries the full set: CAPE sharpened by the lifted index, capped by convective inhibition, then modulated by the 700-500 hPa lapse rate and 0-6 km shear. Shear also decides organisation — a severe-intensity hour with weak shear is reported as "marked, isolated" rather than severe. `capped_gun` flags a loaded-gun setup: lots of fuel held down by a strong cap, explosive if it breaks. ⚠[U+FE0F] This describes an ENVIRONMENT at basin scale (the point is snapped to a ~25 km grid). It does NOT locate individual cells and must never be phrased as "a storm at 4pm" — say "conditions favourable to storms during the afternoon". Convection-resolving models place cells; see `list_models_for_location` for which ones cover this point.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
get_tideHigh and low water times and tidal range at any point on the globe, and real water heights wherever a national almanac covers it.Reads text others writeNo findings
Why it is in this group
- The name "get_tide" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it High and low water times and tidal range at any point on the globe, and real water heights wherever a national almanac covers it. Two very different answers, and the difference matters: - `reference.kind == "chart_datum"` — a national tide almanac is attached to this point (Canada CHS, France api-maree.fr, Ireland Marine Institute). `height_m` is then a TRUE water height above chart datum, which can be added to a charted sounding. - `reference.kind == "lowest_low_water"` — no almanac here, so the figures come from a global model referenced to mean sea level. Times and ranges are sound (they are differences, the unknown datum cancels out) but `height_m` is NOT a water height: it is an offset above an internal reference. Never add it to a sounding, and never present it as depth. Model range is systematically UNDERSTATED, the more so the more confined the place (-6% Halifax, -20% at the head of the Bay of Fundy): 8 km grid cells do not resonate a funnel. This is flagged, never silently corrected. The almanac attachment for a new point is decided in the background, so the first call at a fresh location may return the model and a later one the almanac.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
get_zone_alertsMarine and weather warnings in force for a point, from every official source at once — gale and storm warnings (BMS Météo-France, AEMET, MeteoAM, Met Office) plus European MeteoAla…Reads text others writeNo findings
Why it is in this group
- The name "get_zone_alerts" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Marine and weather warnings in force for a point, from every official source at once — gale and storm warnings (BMS Météo-France, AEMET, MeteoAM, Met Office) plus European MeteoAlarm warnings resolved to the actual region, not the whole country. Empty list means no warning is in force, which is a real answer worth stating. Each warning carries its issuing service, validity window, level and the official text in the publishing language. Check this before any passage or coastal activity. Costs no third-party call when warm (own 15-minute cache).
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
list_models_for_locationWhich weather models actually cover a point, and which resolve convection.Reads text others writeNo findings
Why it is in this group
- The name "list_models_for_location" says "list": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Which weather models actually cover a point, and which resolve convection. A regional high-resolution model hides its global parent where it is available (AROME 1.5 km over France, ICON-2I over the Mediterranean, UKV over the UK, HRDPS over Canada). `resolves_convection` marks the models fine enough to place individual storm cells rather than parametrise them — those are the ones to cite when asked where a storm will actually break. `near_domain_edge` warns that the point sits in a regional model's relaxation zone, where its values are least trustworthy. Costs nothing: read from the registry, no network call.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
- readOnlyHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write9
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:54 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 0
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools1 changed.~1 changedView diff
- Changed:analyze_anchorageOnly readsdescription
Tools1 changed.~1 changedView diff
- Changed:analyze_anchorageOnly readsdescription
Install
Pinned to version 1.2.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http solano https://solanometeo.fr/mcpclaude mcp add --transport http solano https://solanometeo.fr/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__solano__get_forecast",
"mcp__solano__get_confidence",
"mcp__solano__get_storm_risk",
"mcp__solano__get_marine_bulletin",
"mcp__solano__get_zone_alerts",
"mcp__solano__get_tide",
"mcp__solano__find_known_anchorages",
"mcp__solano__analyze_anchorage",
"mcp__solano__list_models_for_location"
],
"ask": [],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as solano. Review it before you use it.
Capability tags behind 9 rules
- mcp__solano__get_forecast untrusted-input
- mcp__solano__get_confidence untrusted-input
- mcp__solano__get_storm_risk untrusted-input
- mcp__solano__get_marine_bulletin untrusted-input
- mcp__solano__get_zone_alerts untrusted-input
- mcp__solano__get_tide untrusted-input
- mcp__solano__find_known_anchorages untrusted-input
- mcp__solano__analyze_anchorage untrusted-input
- mcp__solano__list_models_for_location untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/fr.solanometeo/solano/verify?hash=138eeba97cceb56ae3e3cb72e96dec1d7b3dc01ac10bf044a082223ee95b6a04" | jq -e '.changed == false'Pinned hash 138eeba97cce, from Nov 15, 2026, 10:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/fr.solanometeo/solano/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/fr.solanometeo/solano)Source: MCP Registry. Synced Nov 15, 2026, 10:17 UTC.
Solano — marine & outdoor weather MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Solano — marine & outdoor weather MCP server safe?
- No scan can promise that. What its definitions show: 9 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.
Can the Solano — marine & outdoor weather MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Solano — marine & outdoor weather MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the Solano — marine & outdoor weather MCP server have?
- 9 tools: 9 read. get_forecast, get_confidence, get_storm_risk, get_marine_bulletin, get_zone_alerts, get_tide, find_known_anchorages, analyze_anchorage and 1 more.
Does the Solano — marine & outdoor weather MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Solano — marine & outdoor weather MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Solano — marine & outdoor weather MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 15, 2026. Every change is on this page and in its Atom feed.