| name | info-diet |
| description | Information-diet lens on Apple Screen Time — what you feed your brain, not how long you're on a device. Use for "info diet", "information diet", "what am I consuming", "reading vs feeds", "am I doomscrolling", "attention quality", "is my content junk or nutrition", "social feed time", "news vs deep reading". Mind-area analysis; for total screen/phone volume use the lifestyle `digital-wellbeing` skill instead. |
Information diet (mind lens)
Judges the quality of what your brain consumes, using unified Apple Screen Time
(Mac + iPhone). The question is nutrition vs. junk — intentional/deep intake vs.
passive algorithmic feeds — NOT total hours (that's the lifestyle digital-wellbeing
skill). Same underlying data, different question.
Get the data
Run the bundled extractor (self-contained — Python stdlib + vendored ccl_segb; needs
Full Disk Access), or read the latest vault snapshot if fresh:
python3 "${CLAUDE_PLUGIN_ROOT}/skills/info-diet/screentime_extract.py" --days 30 \
--snapshot-dir "<vault>/400 Resources/screentime"
The extractor is identical to the lifestyle digital-wellbeing copy (App.InFocus SEGB →
per-app/device/day); see that skill's provenance notes for internals.
The lens: categorize by information TYPE
Bucket app time into:
- Algorithmic feeds (junk risk — passive, engineered-for-engagement): Instagram
com.burbn.instagram, X/Twitter, TikTok com.zhiliaoapp.musically, Reddit, Facebook,
YouTube feed com.google.ios.youtube, LinkedIn feed com.linkedin.LinkedIn.
- News/current: news apps, browser news reading.
- Deep / learning / creation (nutrition): Obsidian
md.obsidian, Claude
com.anthropic.claude*, Books/Kindle, Readwise, long-form reading, courses.
- Messaging (communication, not diet): WhatsApp, Telegram
ph.telegra.Telegraph,
Slack com.tinyspeck.chatlyio, Messages.
- Video/entertainment: Netflix, YouTube (lean-back).
Report
- Lead with the feed-junk : deep-intake ratio and the single biggest algorithmic
sink (usually a social feed on iPhone). Trend it across snapshots if history exists.
- Separate iPhone (where feeds dominate) from Mac (where deep work lives).
- Ignore the
(idle/system) bucket — it's not consumption.
- Take a position and recommend for the brain: e.g. "Instagram is 60% of your feed
intake at Xh/wk — cap it, replace with the reading queue." Ground in the
300 Areas/3 Mind protocols. Voice guide applies (concrete, <25-word sentences, end with a rec).
