Control Tesla vehicles from macOS via the Tesla Owner API using teslapy (auth, list cars, status, lock/unlock, climate, charging, location, and extras). Use when you want to check your car state or run safe remote commands. Designed for Parth Maniar (@officialpm) with local-only auth caching, confirmation gates for disruptive actions, and chat-friendly status output.
Security Analysis
medium confidenceThe skill appears to implement the advertised Tesla controls, but there are inconsistencies (undeclared environment requirements, missing install/dependency metadata, and an unknown/undocumented source) that you should resolve before installing.
The code and SKILL.md clearly implement Tesla Owner API functionality via teslapy (auth, list, status, lock/unlock, climate, charging, location, mileage DB, etc.), which matches the skill name and description. However, registry metadata lists no required env vars or primary credential even though SKILL.md and the CLI require TESLA_EMAIL and local OAuth token caching — this metadata omission is inconsistent and may mislead users about what the skill needs.
Runtime instructions are focused on Tesla operations and local-only storage: browser-based OAuth, paste callback URL, and local token cache (~/.tesla_cache.json), defaults (~/.my_tesla.json), and an optional local SQLite DB (~/.my_tesla/mileage.sqlite). Safety gates (--yes) and --no-wake options are present for disruptive actions. No hidden external endpoints are present in SKILL.md; behavior stays within the stated purpose.
There is no install spec (so nothing will be automatically downloaded), but the main script declares a dependency on teslapy in an in-file comment. The registry metadata did not declare this dependency or provide an install step. That mismatch can cause surprise at runtime (missing package), and because code files exist, reviewers should verify how dependencies will be provided before running.
The SKILL.md requires TESLA_EMAIL and optionally MY_TESLA_DEFAULT_CAR, and the CLI creates local OAuth token caches; yet the registry metadata lists no required environment variables or primary credential. The required TESLA_EMAIL (sensitive account identifier) is not declared in the skill metadata, which is an incoherence that impairs informed consent. Other than that, the env/credential needs are proportional to the stated purpose (no unrelated cloud keys or broad credentials are requested).
The skill writes local files in the user's home (~/.tesla_cache.json, ~/.my_tesla.json, ~/.my_tesla/mileage.sqlite) and documents a macOS launchd example to run mileage recording hourly. always:false (no forced inclusion). The files are local-only and the script attempts best-effort chmod 0600. Be aware that if you schedule it or let an autonomous agent invoke the skill, it could perform API calls (safety-gated actions require --yes but an agent could pass that flag).
Guidance
This skill largely does what it says, but there are a few mismatches you should address before installing: 1) The SKILL.md requires TESLA_EMAIL and uses teslapy, but the registry metadata does not declare any required environment variables or a dependency/install step — verify and supply TESLA_EMAIL yourself rather than putting it in a public plist. 2) Review the scripts/tesla.py source (it is included) and confirm you trust the author and the code before running; source/homepage are unknown. 3) Ensure teslapy >=2.0.0 (or appropriate version) is installed in a controlled Python environment (virtualenv) to avoid surprising system-wide installs. 4) Expect local token and defaults files in your home; consider running the first auth interactively and check that ~/.tesla_cache.json and ~/.my_tesla.json have safe permissions. 5) If you plan to allow autonomous agent invocation or schedule the script via launchd, be mindful that remote control commands exist (unlock, charge start/stop, defrost, trunk, windows, etc.) and that an agent or script with the --yes flag could trigger them; keep the skill's usage restricted to trusted agents/users. If you want to proceed, run the included tests locally in an isolated environment (PYTHONDONTWRITEBYTECODE=1 python3 -m unittest discover -s tests) and confirm dependency availability and the expected behavior.
Latest Release
v0.1.64
charge status shows usable battery + charging power details
More by @officialpm
Published by @officialpm on ClawHub
