CommunityActive· 29d★ 4.3by Tunahan Cobanov

TeslaMate MCP

Queries your TeslaMate PostgreSQL database with 18 predefined analytics tools for battery, charging, and driving data.

View repo

Safety & Trust

Delv Safety Grade: C

Score 58/100 · assessed 2026-04-18

Maintainer45

Permissions75

Supply chain40

Transparency70

Incidents100

This community MCP server connects to a TeslaMate PostgreSQL database to query vehicle telemetry. The maintainer appears to be a solo developer with limited GitHub presence. The server requires direct database credentials via TESLAMATE_DB_URL, creating a single point of credential exposure. Installation is manual via uv sync with no package registry distribution, increasing supply chain risk. The repository shows basic documentation and 18 predefined read-only queries for battery health, charging patterns, and driving statistics. Permissions are reasonably scoped to database reads only, which is positive. However, the lack of package distribution, solo maintenance, and requirement for direct database access to potentially sensitive vehicle data warrant caution. No security incidents are known. Suitable for personal use by those comfortable with the technical setup, but organisations should evaluate credential management carefully.

Green flags

Read-only database queries, no write or delete operations
Open source with clear repository and basic documentation
Scoped to specific TeslaMate database schema, not arbitrary SQL
No known security incidents or malicious behaviour
Predefined queries reduce SQL injection surface

Red flags

Solo maintainer with limited public track record
Requires direct PostgreSQL credentials in environment variable
No package registry distribution, manual clone-and-build only
Accesses potentially sensitive vehicle location and usage data
Limited maintenance activity and community review

Permissions requested

DB readOutbound networkAccess secrets

Assessed by Delv Editorial using public metadata. Grades are advisory and update as the ecosystem changes. They do not replace your own review of permissions and code before granting an agent access to sensitive systems.

Install

uv sync

Env vars needed: TESLAMATE_DB_URL

Review

TeslaMate MCP plugs straight into your TeslaMate PostgreSQL database and surfaces 18 ready-made analytics tools for querying battery health, charging sessions, driving patterns, and cost breakdowns. If you're already running TeslaMate (the open-source Tesla data logger), this gives Claude direct access to your vehicle's telemetry without you writing SQL or exporting CSVs. The tools cover the obvious questions: battery degradation over time, charging efficiency by location, cost per kilometre, trip summaries, and idle periods. I'd reach for this when I want to ask "how much did I spend charging at home versus Superchargers last month?" or "what's my battery's state of health trend since purchase?" without opening Grafana. The 18 predefined tools are specific enough to be useful but not so granular that you're drowning in options. You get battery state of health, charging stats by location, driving efficiency, cost analysis, and geofence summaries. The repo lists every tool with its parameters, which is helpful because Claude will need to know what's available. The limitation is that you're locked into these 18 queries. If TeslaMate logs something the tools don't expose, you'll need to fork the server or write raw SQL yourself. Setup assumes you're comfortable with environment variables and have TeslaMate already running with its Postgres instance accessible. The connection string format is standard Postgres, so if you've pointed any other tool at your TeslaMate database, this will feel familiar. The server doesn't touch your car directly; it's read-only against the database TeslaMate maintains. That's a good constraint: no risk of sending commands to the vehicle, but also no live data unless TeslaMate is actively logging. This is a single-purpose tool. If you don't own a Tesla or don't run TeslaMate, it's irrelevant. If you do run TeslaMate and find yourself repeatedly exporting data or writing the same Grafana queries, this is a clean shortcut. The 18 tools won't cover every edge case, but they hit the common analytics questions without making you a SQL expert.

Verdict

Install this if you're already running TeslaMate and want Claude to answer battery, charging, and driving questions without leaving the chat. Skip it if you don't own a Tesla, aren't logging with TeslaMate, or need custom queries beyond the 18 predefined tools.

Good at

Eighteen predefined tools cover the most common EV analytics questions without writing SQL.
Read-only database access means no risk of accidentally sending commands to your vehicle.
Works directly with your existing TeslaMate Postgres instance, no data export or sync required.
Tool descriptions are clear enough that Claude can pick the right query without you specifying parameters every time.

Watch out

Requires TeslaMate already running and accessible, so it's useless if you're not logging Tesla data.
Locked into 18 predefined queries; custom analytics need a fork or raw SQL access.
No live vehicle data, only what TeslaMate has logged to the database.
Setup assumes you're comfortable with Postgres connection strings and environment variables.

Use cases

EV analytics
charging cost analysis
battery degradation tracking
driving pattern insights

Getting started

1. Clone the repo and run `uv sync` to install dependencies. 2. Set the `TESLAMATE_DB_URL` environment variable to your TeslaMate Postgres connection string (format: `postgresql://user:password@host:port/database`). 3. Add the server to your Claude Desktop config under `mcpServers` with the path to the installed server and the environment variable. 4. Restart Claude Desktop and ask it to list available TeslaMate tools to confirm the connection. 5. Watch out: the server is read-only and only as current as your TeslaMate logging. If TeslaMate isn't running or your database is stale, results will lag behind real-time.

Works with

Claude DesktopClaude CodeCursor