Earl Display Control
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill mostly matches its TV-dashboard purpose, but it deserves review because it can serve household dashboard data over an unauthenticated local web server.
Install only if you are comfortable with a local dashboard that stores household state on disk. Before running the server, consider binding it to localhost, for example with `python3 -m http.server 8000 --bind 127.0.0.1`, or using firewall rules. Avoid putting secrets in `earl_mind.json`, use approximate weather coordinates if needed, and be careful with force-kill or clear operations.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Household reminders, room notes, long-term patterns, and configured location could be visible beyond the TV machine depending on the network and firewall.
The documented server command starts a simple unauthenticated HTTP server for the directory containing the dashboard state. Without an explicit localhost bind, Python's server may be reachable from other devices on the network.
`python -m http.server 8000` ... `The dashboard reads earl_mind.json` ... `contains real household state`
Bind the server to 127.0.0.1 or otherwise firewall/authenticate it, and avoid putting sensitive household details in the dashboard state.
Your configured approximate home location leaves the device when the weather helper runs.
The weather helper sends the configured latitude, longitude, timezone, and units to Open-Meteo to fetch weather.
`latitude={_lat}&longitude={_lon}` ... `urllib.request.urlopen(URL, timeout=10)`Use approximate coordinates or leave latitude/longitude at 0.0 if you do not want live weather location data sent to the provider.
A mistaken command could kill another application using port 8000 or wipe local dashboard content.
The skill documents forceful process termination for port 8000 and local content-clearing methods. These are purpose-aligned for fixing the dashboard but can disrupt unrelated local processes or erase dashboard items if used casually.
`lsof -ti:8000 | xargs kill -9` ... `clear_house_stuff()` ... `clear_sketchpad()`
Use the kill and clear operations only when explicitly intended, and prefer graceful shutdown or backups for important dashboard state.
Private notes or inaccurate entries placed in the dashboard state may persist and be reused in future interactions.
The skill persistently stores household patterns and can return the raw mind state for agent context.
`self.mind["long_term_patterns"].append(...)` ... `def snapshot(self):` ... `return self.mind`
Do not store secrets or sensitive household details in Earl's mind file, and periodically review or reset persisted patterns.