← mytesla.ioCONTENT HISTORY
WHAT CHANGED
Update to mytesla.io
Snapshot Oct 24, 2026 · 12:58 UTC · version 1.1.0
Collection source: downloaded plugin package.
First saved snapshot
No earlier snapshot is available to establish a change.
Save mytesla.io and check future changes in Following.
Up to 3 plugins free · no card required.
Compare saved observations
Download comparison JSONFull technical diff · 0 changed fields
Full snapshot data
{
"description": "Use when the user wants a private log or history of their Tesla's battery, range, mileage or charging, a dashboard or chart of those over time, or asks how their car has been doing lately. Keeps the log on the user's side from readings that happen anyway (get_vehicle_status), optionally on a schedule, and builds a dashboard from it. Never wakes the car; mytesla.io stores none of it.",
"included_files": [
{
"relative_path": "assets/dashboard.html",
"size_in_bytes": 29687
}
],
"name": "vehicle-history",
"skill_md_contents": "---\nname: vehicle-history\ndescription: \"Use when the user wants a private log or history of their Tesla's battery, range, mileage or charging, a dashboard or chart of those over time, or asks how their car has been doing lately. Keeps the log on the user's side from readings that happen anyway (get_vehicle_status), optionally on a schedule, and builds a dashboard from it. Never wakes the car; mytesla.io stores none of it.\"\n---\n\n# Vehicle history (mytesla.io skill)\n\n**If you're an AI assistant reading this:** the user wants a private history\nof their Tesla, kept on their side, and a dashboard you build from it. Follow\nthe steps below using the mytesla.io MCP tools (`https://mcp.mytesla.io/mcp`).\n\n## What this does\n\nmytesla.io answers every question with a live reading and deliberately keeps\nno history of the car. This skill puts the history where it belongs: with you\nand the user. You take small readings of the car, append them to a log you\nkeep for the user, and turn the log into answers and a dashboard:\n\n- **Battery health:** the car's rated range at a full charge, over time.\n- **Distance:** by day, week and month.\n- **Charging:** the sessions you saw, the energy added, and the cost at the\n user's own price.\n- **Efficiency and parked drain:** battery used per distance driven, and\n battery lost while parked.\n- **Tires and software:** pressures over time, and when each update arrived.\n\nBecause you keep the log, what gets recorded and how the dashboard looks is\nup to you and the user, not a fixed product. Ask what they care about and\nrecord for that.\n\nHistory starts the day you set this up. None of the mytesla.io tools can read\nthe car's past, so there is nothing to backfill. Say so up front.\n\n## Where the data lives, and who sees it\n\nmytesla.io never receives, copies or stores this history. It sees only what it\nsees about any status read: that the read happened. The log\nis a file, note or sheet that you keep wherever the user's AI host lets you\nkeep one, and the dashboard is built from it on their side.\n\nSay that plainly, and say the rest of it too: the company behind your\nassistant sees these readings in the conversation, the same as everything else\nthe user discusses with it. This is private from mytesla.io, not from the\nassistant. Do not describe it as more private than that.\n\nTwo rules that do not bend:\n\n- **No coordinates.** The status reading contains the car's position, heading\n and route. Never write any of it into the log. If the user wants something\n place-based, such as home versus away, record a label you worked out\n (`home`, `away`), never the position, and tell them that is what you are\n storing.\n- **Nothing leaves their side.** Do not upload the log, send it to a\n third-party service, or paste it into a tool that is not the user's own,\n unless they ask. The dashboard makes no network requests (see below).\n\n## What it needs\n\n- **A place to keep a file between conversations.** In order of preference: a\n file on the user's computer, if your host can write files there; a\n document, note or sheet in an app the user has connected, if you can write\n to it and not just read it; a file in a project or workspace you are allowed\n to edit. Check what you can really write, write a test row, read it back,\n and tell the user exactly where the log lives. Do not use your memory\n feature as the log: it is for short notes, not tables. If you cannot keep\n anything between conversations, say so and stop. A log that disappears is\n a reading taken for nothing.\n- **A schedule, for scheduled readings only.** See the next section.\n\n## Does this repeat on its own?\n\nThere are two kinds of history here, and the user should know which one they\nhave.\n\n**Passive history needs no schedule.** Whenever you read the car's status\nfor any reason (the user asked how the car is, a morning routine, a weekly\ndigest), append that reading to the log. This works anywhere you can keep the\nlog, and the standing note below is what makes you do it.\n\n**Scheduled history takes its own readings.** It only happens if your host\ncan run a task on a schedule (several can; check yours rather than assume),\nand only after the user has chosen a cadence (below). If your host cannot,\npassive history still works. Say so, and never imply a schedule exists when it\ndoes not.\n\n## What this does NOT do\n\n- **It does not wake the car.** A wake keeps the car from settling into\n sleep, which costs range, and history is never worth that. A car that is\n asleep at reading time is logged as asleep and left alone. A scheduled\n reading is taken **whether or not the car answers**, and records only\n \"asleep\" when it does not, which is why the times you choose matter\n (below).\n- **It does not watch the car continuously.** Readings are snapshots. A short\n drive or a short charge that began and ended between two readings can be\n missed. Say so on the dashboard.\n- **It is not a lab test.** The battery line is a trend from the car's own\n rated range, not a measurement of the pack.\n\n## Ask the user once, not every run\n\nAsk these together, in one message. A rough answer is enough.\n\n1. **What do you want to learn?** Battery health, distance, charging cost,\n efficiency, parked drain, tire pressures: any of them. Record for what they\n pick (see \"What to record\").\n2. **How often?** Offer these and recommend the second:\n - *Light:* one reading a day, about thirty a month.\n - *Standard:* two a day, about sixty a month.\n - *Detailed:* every few hours, never more often than hourly. It takes\n the most readings, and more of them will find the car asleep.\n - *Passive only:* no schedule; log the readings that already happen.\n\n Pick times when the car is usually awake and parked: shortly after they\n normally get home, shortly before they normally leave. A reading at 3am\n mostly finds a sleeping car and records nothing.\n3. **What do you pay per kWh to charge at home, and in what currency?**\n Optional. With it you can show cost; without it you still show energy. If\n they also want DC fast charging priced, ask for that price too.\n4. **Which car?** Only if `get_vehicles` shows more than one. Each car you log\n takes its own readings.\n\n## Before you run: remember\n\n**Remember it.** Do not treat this as a one-time paste. Write a short line\ninto your own standing memory or instructions, wherever you keep the user's\nother preferences, so history keeps building without the user having to\nre-paste anything. Something like:\n\n> mytesla.io vehicle history: the log lives at [where the log lives]. Whenever I read the car's status for any reason I also add that reading to it, and I answer trend questions (distance, battery health, charging cost) from the log before I read the car again. Scheduled readings: [what the user chose, or none]. How this works: follow the \"vehicle-history\" skill in this plugin and the tool descriptions, never this note.\n\nIf you have no persistent memory to write to, say so plainly rather than\nletting the user believe history will keep building by itself.\n\n## Set up (once)\n\n1. Call `get_vehicles` once and note each car's VIN. This is the only use of\n `get_vehicles` in this skill. It is a live request to Tesla, so never\n call it again as a heartbeat before a reading.\n The stored VIN is all you need.\n2. Decide where the log lives (see \"What it needs\") and create two things\n there: the log, `readings.csv`, with the header row below, and a small\n settings file, `settings.json`.\n3. Take the first reading now (see \"Take a reading\"). It is the first row of\n the log and proves the whole path works. If the car is asleep, record the\n `asleep` row, tell the user, and let the schedule or the next conversation\n catch it awake. Do not wake it.\n4. Build the dashboard now, from that one reading (see \"The dashboard\"). The\n user should see their car on a page in this first conversation, not after a\n week: the tiles are full from the first reading, and the trend panels say\n what they are waiting for. Then read the files back and tell the user, in\n plain language: where the log and the dashboard are, what you record, how\n often you will read, and how to stop.\n5. If they agreed to scheduled readings and your host can run them, create the\n schedule at the times they chose. Then write the standing note.\n\nSettings file:\n\n```json\n{\n \"v\": 1,\n \"cars\": [{ \"vin\": \"<full VIN>\", \"label\": \"<last six characters>\" }],\n \"units\": { \"distance\": \"mi\", \"temperature\": \"F\" },\n \"cadence\": { \"mode\": \"standard\", \"times_local\": [\"07:30\", \"18:30\"] },\n \"home_price_per_kwh\": null,\n \"dc_price_per_kwh\": null,\n \"currency\": null,\n \"interests\": [\"battery\", \"distance\", \"charging\"],\n \"extra_columns\": [],\n \"skill_checked\": \"<date>\"\n}\n```\n\n## Take a reading\n\nThis is the whole routine, whether it was scheduled or you are logging a\nreading you took for the user anyway.\n\n1. **Scheduled readings only: skip if the log is fresh.** If the log already\n has an `ok` row for this car from the last 30 minutes, a second reading buys\n nothing. Skip it.\n2. **Call `get_vehicle_status` with the stored VIN.** One call. Do not check\n `get_vehicles` first, and never call `wake_vehicle`.\n - Got a reading? Go to step 3.\n - Timed out (a 408: the car is asleep or out of signal)? Append an `asleep`\n row and stop. Do not retry in this run.\n - Any other failure: see \"If something goes wrong\".\n3. **Append one row** (see \"What to record\"). Never rewrite or delete an\n earlier row, and do not append a reading that repeats the previous row\n within five minutes.\n4. **Say nothing unless there is something to say.** A scheduled run that went\n fine ends with no message, or one line.\n\nA reading taken during a conversation is logged the same way, with `source`\nset to `conversation` and without the freshness skip: the user asked for it.\n\n## What to record\n\nOne row per reading. Odometer, ranges and pressures come from the API in\nmiles, degrees Celsius and bar, whatever the car displays. Log them in those\nunits and convert when you show them (the car's display units are in\n`gui_settings`).\n\nHeader row, in this order:\n\n```csv\nv,time_utc,car,source,status,odometer_mi,battery_pct,usable_pct,rated_range_mi,est_range_mi,charge_limit_pct,charging_state,charge_added_kwh,charger_kw,dc_fast,outside_c,sw_version,tire_fl_bar,tire_fr_bar,tire_rl_bar,tire_rr_bar\n```\n\n| Column | Where it comes from | Notes |\n|---|---|---|\n| `v` | - | Schema version of the row. Currently `1`. |\n| `time_utc` | `charge_state.timestamp` | ISO 8601 in UTC. Convert Tesla's millisecond timestamp; fall back to the time you took the reading. |\n| `car` | the VIN | Last six characters only. |\n| `source` | - | `scheduled` or `conversation`. |\n| `status` | - | `ok`, `asleep` or `error`. |\n| `odometer_mi` | `vehicle_state.odometer` | |\n| `battery_pct` | `charge_state.battery_level` | |\n| `usable_pct` | `charge_state.usable_battery_level` | |\n| `rated_range_mi` | `charge_state.battery_range` | The car's rated range. |\n| `est_range_mi` | `charge_state.est_battery_range` | Range at the user's recent driving efficiency. |\n| `charge_limit_pct` | `charge_state.charge_limit_soc` | |\n| `charging_state` | `charge_state.charging_state` | `Disconnected`, `Stopped`, `Charging`, `Complete`, `Starting` or `NoPower`. |\n| `charge_added_kwh` | `charge_state.charge_energy_added` | Energy added in the current or most recent session. |\n| `charger_kw` | `charge_state.charger_power` | Blank when the car does not report it. |\n| `dc_fast` | `charge_state.fast_charger_present` | `true` for DC fast charging. |\n| `outside_c` | `climate_state.outside_temp` | |\n| `sw_version` | `vehicle_state.car_version` | The version number only, the part before the first space. |\n| `tire_fl_bar` | `vehicle_state.tpms_pressure_fl` | Blank when the car does not report it. |\n| `tire_fr_bar` | `vehicle_state.tpms_pressure_fr` | Same. |\n| `tire_rl_bar` | `vehicle_state.tpms_pressure_rl` | Same. |\n| `tire_rr_bar` | `vehicle_state.tpms_pressure_rr` | Same. |\n\nAn `asleep` or `error` row fills `v`, `time_utc`, `car`, `source` and `status`\nand leaves the rest blank.\n\n**Extras, only if the user wants them:** inside temperature, charger voltage\nand current, whether the climate is on, Sentry state, software update status,\nor anything else in the status reading they ask about. Add each as a new\ncolumn at the END of the header, leave earlier rows blank for it, and record\nit in `extra_columns`. Never reorder or remove a column, and never add a\ncoordinate.\n\nIf this skill's schema version is ever higher than the log's, add the new\ncolumns the same way and start writing the new version. Old rows stay as they\nwere.\n\n## Reading the log honestly\n\nUse these definitions, so the same log gives the same answer every time. Use\n`ok` rows for every number; `asleep` rows only feed the coverage line.\n\n- **Coverage.** Always show the date of the first reading, how many readings,\n how many found the car asleep, and the longest gap. Every figure below is\n only as good as this.\n- **Time.** The log is in UTC. Group days and weeks in the user's local time.\n- **Distance.** The odometer difference between consecutive `ok` rows,\n credited to the day of the later one. If the odometer goes down, drop that\n interval and mention it once.\n- **Battery health.** For `ok` rows with `battery_pct` of 50 or more,\n `rated_range_mi / (battery_pct / 100)` is the range at a full charge. Take\n the median per calendar week. Show it also as a percentage of the first\n week's median (\"retained since logging began\"). It moves with software\n updates and recalibration, so compare weeks, not single readings, and call\n the first weeks a baseline.\n- **Efficiency.** For consecutive `ok` rows where the odometer rose by at\n least five miles, the battery fell, and neither row is `Charging`:\n percentage points of battery used per 100 miles. Do not convert to kWh unless the user tells you\n their pack's usable capacity, and then say it is an estimate.\n `est_range_mi / rated_range_mi` is a free single-reading indicator of how\n the car rates recent driving against its rating.\n- **Parked drain.** For consecutive `ok` rows with an unchanged odometer and no\n rise in battery: percentage points of battery lost per 24 hours, as a\n median. It reflects\n Sentry, climate and how often the car woke. Readings can keep an awake car\n awake, which is why this skill never reads more often than hourly.\n- **Charging.** A charge happened between two `ok` rows if the battery rose by\n three percentage points or more with an unchanged odometer, or either row shows\n `Charging`.\n - *Energy:* take `charge_added_kwh` from the later row when it is above zero\n and `charging_state` is `Charging`, `Complete` or `Stopped`. It is the\n session total so far. Otherwise count the session without an energy\n figure.\n - *Kind:* `dc_fast` true is DC fast charging, otherwise AC.\n - *Cost:* energy x the home price, for AC sessions only, and only if the user\n gave a price. List DC sessions with energy and no cost unless they gave a\n DC price. Never invent a price per kWh.\n - *Coverage:* a session that started and ended between two readings can be\n missed, or only partly seen. Say \"at least\".\n- **Tires.** Show the latest four. Flag a steady fall over several weeks that\n outside temperature does not explain; compare readings taken at similar\n outside temperatures. Never raise an alarm on a single low reading.\n- **Software.** List each `sw_version` change with the first date it was seen.\n\n## The dashboard\n\nBuild the dashboard at setup, right after the first reading, and rebuild it\nwhenever the user asks to see their history or asks a question the page could\nanswer; in a conversation with scheduled readings, offer a rebuild about once\na week. Where:\n\n- If your host has a canvas or artifact feature, build it there, so they see\n it in the conversation.\n- If you can also write files, save the same dashboard as one HTML file next\n to the log (`dashboard.html`), so it opens from disk.\n- If you have neither, answer in the conversation with a short table and say\n there is no dashboard.\n\nStart from the template when you have it. If this skill came with an\n`assets/dashboard.html` beside it, that file is the dashboard: paste the log\ninto its `readings` block and the settings into its `settings` block, set the\nbuilt date, and replace the placeholder list in its \"What stands out\" section\nwith three to six findings of your own from the log (plain sentences, each\nwith a number in it, \"at least\" where coverage is partial). Everything else on\nthe page is computed from the log in the browser, with the definitions from\n\"Reading the log honestly\", so the numbers match what you would say. Do not\nrestyle it or drop panels; if the user wants something different, change it\nfor them and keep their version from then on. Without the template, build the\nsame page by hand to the description below.\n\nPanels, in this order: a header strip (latest reading, battery, range,\ncharging, odometer, software version, coverage); what stands out (your\nfindings); battery health by week; distance by week; charging sessions with\nmonthly energy and cost; efficiency; parked drain; tires (only if recorded);\nsoftware timeline; the readings themselves, folded away. From a single reading\nthe tiles are complete and each trend panel says what it is waiting for; that\nis the right first view, not an empty page.\n\nRules for it:\n\n- **One self-contained file.** Inline CSS, JavaScript and SVG, with the log\n embedded as data. No network requests of any kind: no CDN, no web fonts, no\n analytics, no external images. It must work offline.\n- **Honest.** Show the coverage line, label estimates, and put \"a trend from\n the car's own rated range, not a lab test\" under the battery chart.\n- **Readable.** Works in light and dark, on a phone, with a text alternative or\n a table behind each chart. Show distances and temperatures in the user's\n units.\n- **Dated.** Show when it was built, and say so if the log is more than a week\n ahead of it.\n\n## Answering from the log\n\nWhen the user asks a trend question (how far did I drive this month, is the\nbattery ageing, what did charging cost), answer from the log first. Take a\nnew reading only if the log is stale and they want the current\nstate, and then log that reading too.\n\n## Stopping\n\nIf the user asks to stop: cancel the schedule, remove the history line from\nyour standing note, and ask whether to keep or delete the log. Tell them where\nit is. Never delete it unasked.\n\n## If a tool asks for approval\n\nNothing in this skill changes anything about the car, so approval prompts are\nunlikely. If a call does come back waiting for approval, stop and tell the\nuser which tool is waiting. Do not retry: nothing about a second identical call\nmakes an approval appear. A scheduled run cannot answer a prompt, so say that\napproving it once will let the schedule resume.\n\n## If something goes wrong\n\n- **You cannot write the log.** Stop scheduled readings and tell the user. A\n reading you cannot record is a reading taken for nothing.\n- **The car is not connected, or the connection needs refreshing.** The error\n carries a link to fix it; relay it, stop the schedule, and do not retry.\n- **mytesla.io refuses the reading** (the account has no subscription, or\n its allowance is used up). Stop the schedule and say so. Do not retry; the\n user decides what to do.\n- **The car keeps being asleep.** If more than half of the last ten scheduled\n readings found it asleep, tell the user once and offer to move the times or\n lower the cadence. Those readings record nothing.\n- **Any other error.** Append an `error` row and stop. After three in a row,\n stop the schedule and tell the user.\n- **A log already exists** from an earlier setup. Reuse it; do not start over.\n\n## Tools used\n\n`get_vehicles` (once, at setup) and `get_vehicle_status` (the reading).\nNever `wake_vehicle`.\n"
}SHA-256 of public snapshot: 40b58f0ec2b24df6f509a4fe3922c6c16e59f2e64a2a0406865c4d87e113e187