Comprehensive Model Context Protocol server for managing Home Assistant through AI assistants.
Nothing to run from here
durely insights MCP for Garmin 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.
97/100
trust score
17
downloads a day
local
runs on your machine
820
downloads in 30d
What it does
Your full Garmin history, kept locally, analysed down to the second, via the official Personal API.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user durely-insights-mcp-for-garmin '--' npx -y @durely/garmin-insights-mcp
# 2 — ask your agent for something these tools do
# garmin_analyze_activity, garmin_training_summary, garmin_recovery_trends and 18 more
durely insights MCP for Garmin costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
DUAll 1 server
durely1 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 runs on your machine: your client starts durely insights MCP for Garmin 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.
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. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user durely-insights-mcp-for-garmin '--' npx -y @durely/garmin-insights-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free 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
Nothing is required, but it can take GARMIN_PERSONAL_API_KEY (Your Garmin Personal API key. Leave empty if you saved it with `npx -y @durely/garmin-insights-mcp auth` (recommended): the server then reads it from ~/.config/garmin/connect-personal-api-key.), GARMIN_MCP_TIMEZONE (Your time zone, e.g. Europe/Brussels. Defaults to your computer's time zone.), GARMIN_API_KEY_EXPIRES (The date your key expires, e.g. 2027-10-24, so the server can warn you in time.) if you want to set them.
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
21 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.
Compute training metrics for one activity from its second-by-second FIT data: - time in heart rate zones (the user's Garmin zones for that sport) and, with an FTP, in power zones (Coggan) - power: average, normalized power (NP), variability index, intensity factor and TSS (needs FTP) - aerobic…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| activity_id | integer | yes | Garmin activity ID from garmin_list_activities |
| ftp_w | integer | no | Functional threshold power in watts, for power zones, IF and TSS |
| response_format | yes | — |
Summarize training per week or month: sessions, hours, distance, elevation and training load, per sport; intensity distribution (share of time in low = HR zones 1–2, moderate = zone 3, high = zones 4–5, with a rough pattern: polarized, pyramidal, threshold-heavy); and the load trend at the end of…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Local calendar date, yyyy-MM-dd |
| end_date | string | no | Defaults to today |
| group_by | string | no | — |
| sport | string | no | Only this sport (parent type) or activity type |
| response_format | yes | — |
Relate training load to recovery markers, day by day: resting HR, overnight HRV, sleep score and duration, Body Battery max and stress, each against the user's own rolling 28-day baseline. Returns: a status (last 7 days vs the previous 28: HRV, resting HR, sleep, load; plus current…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Local calendar date, yyyy-MM-dd |
| end_date | string | no | Defaults to today |
| response_format | yes | — |
Compare 2–6 activities side by side: date, type, duration, distance, pace/speed, heart rate, power (avg/NP), elevation, training effect and load, cadence, and efficiency factor (speed or NP per heartbeat; higher means more output for the same heart rate, a classic way to track aerobic progress on…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| activity_ids | array | yes | — |
| deep | boolean | no | Also run FIT-based analysis for each activity |
| response_format | yes | — |
Run one read-only SQLite query against the local Garmin cache, for analyses the other tools don't cover (custom aggregations, correlations, records, comparisons across years). … Rules: one SELECT or WITH statement; at most max_rows rows returned (default 200); 10-second timeout. Prefer…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sql | string | yes | A single SELECT or WITH query |
| max_rows | integer | no | — |
| response_format | string | no | — |
Get day-by-day wellness and recovery data, from the local cache (all history). Per day: steps (+ goal), floors, resting/min/max heart rate, stress (average, max, % of the day at rest/low/medium/high), Body Battery (min, max, charged, drained), intensity minutes (moderate, vigorous), calories…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Local calendar date, yyyy-MM-dd |
| end_date | string | no | Defaults to today |
| group_by | string | no | — |
| response_format | yes | — |
Get nightly sleep from the local cache: bed and wake time, total sleep, deep/light/REM/awake time, naps, sleep score with quality and Garmin's feedback, awake count, average respiration and sleep stress. JSON mode adds the sleep score's components (duration, stress, deep/light/REM %, restlessness,…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Local calendar date, yyyy-MM-dd |
| end_date | string | no | Defaults to today |
| response_format | yes | — |
Get nightly HRV (heart rate variability, in ms) from the local cache: last-night average, highest 5-minute value, 7-day average, the personal baseline range (balanced low/high) and Garmin's status (BALANCED, UNBALANCED, LOW, POOR). With readings_date, also returns that night's 5-minute HRV…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Local calendar date, yyyy-MM-dd |
| end_date | string | no | Defaults to today |
| readings_date | string | no | Night (wake-up date) to include 5-minute readings for |
| response_format | yes | — |
Get one day's health data through the day, averaged into intervals: heart rate (bpm), Body Battery (0-100), overnight HRV readings (ms) and SpO2 (pulse ox, %). Returns per metric: min/avg/max, the day's summary values (e.g. resting HR, Body Battery charged/drained, SpO2 averages), and a columnar…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| date | string | no | Defaults to today |
| metrics | array | no | — |
| interval_min | integer | no | Minutes per point |
| response_format | yes | — |
Get weigh-ins from the local cache: weight (kg), BMI, body fat %, body water %, muscle and bone mass (kg), visceral fat, and where each came from (scale, manual, linked app). Includes the change over the period and the weight goal if one is set. Only days with a weigh-in appear. Defaults to the…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Local calendar date, yyyy-MM-dd |
| end_date | string | no | Defaults to today |
| response_format | yes | — |
Get food logged in Garmin Connect (or a linked app such as MyFitnessPal) per day: calories, carbs, protein and fat against the day's goals, and meals; plus hydration (ml) from the daily summary. Food logs are fetched from Garmin on demand and cached. Defaults to the last 7 days, at most 31 per…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Local calendar date, yyyy-MM-dd |
| end_date | string | no | Defaults to today |
| response_format | yes | — |
Get menstrual cycle tracking from Garmin Connect for a date: cycle day, current phase and days until the next, period and cycle length, predicted next cycle, fertile window and ovulation date, plus that day's logged symptoms and notes (day view). Fetched live from Garmin. Returns a clear "not…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| date | string | no | Defaults to today |
| response_format | yes | — |
Search the user's Garmin activities (all history, from the local cache), newest first by default. Returns one page of compact summaries plus the total number of matches: id, date, type, name, duration, distance, pace/speed, HR, power, elevation, training effect, training load, TSS/IF and VO2max…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| start_date | string | no | Only activities on or after this local date, yyyy-MM-dd |
| end_date | string | no | Only activities on or before this local date, yyyy-MM-dd |
| activity_type | string | no | — |
| search | string | no | Text to match in the activity name (case-insensitive) |
| min_distance_km | number | no | Minimum distance in km |
| max_distance_km | number | no | Maximum distance in km |
| min_duration_min | number | no | Minimum duration in minutes |
| max_duration_min | number | no | Maximum duration in minutes |
| include_multisport_legs | boolean | no | — |
| sort_by | string | no | Sort field |
| sort_order | string | no | desc = newest/largest first |
| limit | integer | no | Maximum activities to return (1-100) |
| offset | integer | no | Number of activities to skip, for pagination |
| response_format | yes | — |
Get full details for one Garmin activity by ID. Returns: name, description, type, start time and time zone, device, summary metrics (duration, distance, speeds, HR min/avg/max, power avg/max/NP/20-min, FTP, cadence, stride, calories, training effect + message, training load, TSS, IF, temperature,…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| activity_id | integer | yes | Garmin activity ID from garmin_list_activities, e.g. 12345678901 |
| response_format | yes | — |
Get the laps (and sessions, swim lengths and Garmin's physiological metrics) of one activity, from its original FIT file. Returns per lap: start offset, duration (elapsed and timer), distance, pace/speed, avg/max HR, avg/max/normalized power, cadence, ascent/descent, intensity…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| activity_id | integer | yes | Garmin activity ID from garmin_list_activities |
| lap_offset | integer | no | Skip this many laps (for activities with very many laps) |
| response_format | yes | — |
Get second-by-second data from one activity's FIT file, downsampled to fit in context. Returns per field min/avg/max over the window, plus columnar series: t (seconds from the start of the activity, or of the leg for a multisport leg) and one array per field. Units: heart_rate bpm, power W, speed…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| activity_id | integer | yes | Garmin activity ID from garmin_list_activities |
| fields | array | no | Fields to return: …, or a dev_* field |
| resolution_s | integer | no | Seconds per point (minimum) |
| max_points | integer | no | Upper bound on points returned |
| start_s | number | no | Window start, seconds from the start |
| end_s | number | no | Window end, seconds from the start |
| response_format | string | no | 'json' (default) for the full series, 'markdown' for stats and a short preview table |
Save the original FIT file of an activity to the server's exports folder and return its path, e.g. to import it into another tool (Intervals.icu, Golden Cheetah, a spreadsheet converter). The file is exactly what the device recorded. Saved as <data dir>/exports/<activity_id>.fit; an existing…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| activity_id | integer | yes | Garmin activity ID from garmin_list_activities |
Get the user's Garmin profile settings, heart rate zones per sport, and fitness age. Returns: units and locale, time zone, preferred training days, HR zone floors per sport (with the max HR / LTHR / resting HR used to compute them), and the latest fitness age. Use this before interpreting heart…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| response_format | yes | — |
Check that the Garmin Personal API connection works and report its configuration. Reports: where the API key comes from, whether a token could be obtained and when it expires, the API key's expiry (if GARMIN_API_KEY_EXPIRES is set), the time zone used for calendar dates, and the server version.…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| response_format | yes | — |
Start downloading Garmin data into the local cache. Runs in the background and returns immediately; follow progress with garmin_sync_status. You rarely need this: other tools sync automatically on first use and when the cache is over an hour old. Use it to: - fetch the latest data right now…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| mode | string | no | — |
| from | string | no | — |
| domains | array | no | Only sync these domains. Defaults to all except "fit_files". |
Show what's in the local Garmin cache and the progress of the current or last sync. Returns: the running/last sync (progress, errors), and per data domain the date range covered, number of days, whether all history is in, and the last successful sync.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| response_format | yes | — |
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under Apache-2.0
- 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 durely insights MCP for Garmin 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 97 / 100
Release 0.5.2 · scanned 30 Oct 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 score97
Findings0 threat · 1 capability
Capabilitymoderate
Coveragesource
Verificationsource
Surface digest6be1a02fc57cd5c62ae59240008ccc99036f04a7fadb93d9caefec51aa142105
Score vectorthreat 100 → client 97
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | moderate | capability blast radius (moderate) — client exposure if the model is manipulated | −3 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| 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 surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (21 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. moderate is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| garmin_list_activities | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_activity | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_analyze_activity | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_training_summary | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_recovery_trends | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_compare_activities | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_query | no tags | |
| garmin_get_activity_laps | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_activity_streams | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_download_activity_fit | untrusted-inputexternal-sink | untrusted-input: keyword "download"; external-sink: declares readOnlyHint: false, openWorldHint: true |
| garmin_get_daily_health | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_sleep | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_hrv | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_intraday | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_body_composition | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_nutrition | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_womens_health | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_get_profile | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_status | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| garmin_sync | untrusted-inputexternal-sink | untrusted-input: keyword "fetch"; external-sink: declares readOnlyHint: false, openWorldHint: true |
| garmin_sync_status | 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 digest6be1a02fc57cd5c6
Scanned2026-10-30
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 0.5.2. Pin a release below and the command asks for exactly that one.
0.5.2
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 durely insights MCP for Garmin 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.