3k Downloads
Overview
Provide natural-language control and monitoring of Tesla vehicles via the Tessie API, including battery, climate, charging, drives, location, and basic vehicle state/locks.
Key Advantages
1.Deep Tesla integration via Tessie, leveraging an established third-party platform with rich data and analytics.
2.Covers both monitoring (battery, range, location, drives) and active control (climate, charging, locks, trunks, windows, sentry).
3.Simple configuration using a single Tessie API key; vehicle VIN/ID are auto-detected from the API, minimizing setup friction.
4.Clear, conversational command patterns (e.g., “tessie battery”, “tessie preheat 72”, “tessie start charging”).
5.Explicit use of Tessie’s documented API endpoints and commands, which improves predictability and maintainability.
Use Cases
- Remote preconditioning of the vehicle cabin before departure (preheat/precool, defrost).
- Checking battery state of charge and estimated range via chat before planning trips.
- Starting or stopping charging sessions and adjusting charge limits while away from the vehicle.
- Querying recent drives for distance, energy usage, and locations for personal tracking or simple analytics.
- Checking live vehicle status: locked/unlocked, online/offline, charging status, sleep state, and sentry mode control (if added).
Evaluation Scores
7.5
/ 10
Reliability
7.0
Functionality
8.8
Usability
8.0
Safety
5.5
Performance
7.5
Compatibility
8.5
Based on 1 evaluation · Latest: 3/19/2026
Download Trend
Loading...
Evaluation History (1)
7.5/103/19/2026▼
OS: darwin-x64LLM: anthropic/claude-sonnet-4.5
**Judgment:** High-utility but high-sensitivity control skill for Tesla owners using Tessie, suitable for power users who understand the implications of remote vehicle control and are comfortable granting an API key. Not yet deployed, so it should be treated as in-review and tested cautiously before production use.
**What it does well:**
- Offers broad control and monitoring of Tesla vehicles: battery, range, climate, charging, location, and basic security-related states (lock/unlock, windows, trunks, sentry via Tessie commands).
- Leverages Tessie’s mature API, avoiding direct low-level Tesla API handling and benefiting from Tessie’s richer data (drives, idles, analytics context).
- Configuration is straightforward: a single Tessie API key via config or environment variable, with automatic vehicle/VIN detection.
- Natural language command set is simple and discoverable (e.g., “tessie battery”, “tessie recent drives”, “tessie set charge limit to 90%”).
**Key risks and concerns:**
- **Security & privacy risk:** Any party with access to the assistant instance using this skill could obtain sensitive data (real-time location, recent drives) and issue control commands (unlock, open windows/trunks, adjust charging). Protecting the Tessie API key and access to the assistant is critical.
- **Physical-world impact:** Unlocking the car, opening windows/trunks, or modifying charging behavior are real-world actions; misinterpretation of user intent or prompt injection could have safety, theft, or nuisance implications.
- **Third-party dependency:** Reliability and latency depend on Tessie’s service availability and Tesla’s connectivity. Outages or vehicle sleep states may cause intermittent failures.
- **Not yet deployed:** The metadata notes “NOT YET DEPLOYED – pending review,” indicating it may still need validation (error handling, safe defaults, confirmation flows for high-impact actions).
**Recommended scenarios:**
- Private or tightly access-controlled deployments where the assistant is used by the vehicle owner (or explicitly trusted users) only.
- Users who already rely on Tessie and understand its security model, and who want a conversational interface for frequent tasks like preconditioning and charging management.
- Non-critical convenience use: checking status, range, or drives, where errors are low-impact.
**Caution / when to avoid or restrict:**
- Multi-user or shared-chat environments where it is difficult to ensure that only authorized individuals can issue commands.
- Any deployment lacking strong authentication, access control, and logging around high-impact commands (unlock, open trunks/windows, enable/disable sentry, etc.).
- Automated or agentic workflows that might issue commands without explicit, recent, user confirmation; these should be restricted or require confirmations before executing sensitive actions.
Comments (0)
No comments yet. Be the first!