Bratha

Part 5.0 — The Tactical Toolkit

9 min read

This is Part 5 of 7 in the Productivity Enhancement Series

Table of Contents

The literacy layer, supercharged

This is the layer the old pyramid called “digital fluency”: keyboard shortcuts, email, browsing, calendars, reading, passwords. The unglamorous literacy of using a computer well. It has the highest hourly return of anything in the stack and the lowest status, which is why most people skip it. Everything here was true before AI; AI just moved the ceiling on each one. This article is a tour of the tactics, each with its fundamental and its multiplier. Pick the two or three where you’re weakest; you don’t need all of them at once.

AI as the default interface

The biggest shift in this whole layer is a change of default. The old default for “I need to do a thing on my computer” was: open the app, navigate its menus, do it by hand. The new default is: describe the outcome to an AI assistant and let it do or draft the thing.

This reframes the entire toolkit. You don’t need to master every app’s menus if you can describe what you want. But (and this is the through-line of the series) the assistant is only as good as the context and instructions you’ve given it and the knowledge it can reach. The tactics below are mostly “here’s the fundamental, and here’s how AI now does the heavy part.”

Email: inbox zero, finally easy

The fundamental hasn’t changed: inbox zero means your inbox is a processing queue, not a storage unit. You capture, clarify, and file each message to zero, rather than letting 4,000 emails sit as a guilt pile. What changed is that the processing used to be slow and manual, and now it isn’t.

  • Triage on a schedule. Connect an AI assistant to your email and have it run a daily or weekly pass: summarise what came in, surface the few that actually need you, and draft replies to the routine ones for your approval. You review a short digest instead of scrolling a long inbox. (This is a scheduled task, previewed.)
  • Kill the source of noise. Have it find recurring newsletters and promotions you never open and unsubscribe in a batch. ==Inbox zero is far easier when the inbox stops filling with noise in the first place.==
  • The rule still holds: the inbox is not a to-do list. Real tasks get captured into your workflow; the email itself gets filed or archived.

Keyboard shortcuts and the speed of thought

Every time your hand leaves the keyboard for the mouse, you pay a small tax and a small attention break. Across a day it’s large. The fundamental: learn the shortcuts for the handful of actions you do hundreds of times a day (switch app, switch tab, search, copy/paste-without-formatting, screenshot, new note). A few hours of deliberate practice here pays back for years.

The modern addition is command palettes (the Cmd/Ctrl-K style “type what you want” bar now in most good software) and text expanders (short codes that expand into boilerplate you type constantly: your address, an email template, a code snippet). Both collapse “navigate to the thing” into “name the thing,” which is the same move as AI-as-interface, one level down.

Calendars that schedule themselves

The fundamental from Part 3.0: the calendar is for time-specific commitments only, and it should be one calendar you trust, not five you ignore. The multiplier is scheduling assistance: AI that looks at your commitments and proposes where a task or meeting fits, handles the back-and-forth of finding a slot, and protects your deep-work blocks instead of letting meetings colonise the whole day. You keep the veto; it does the Tetris.

Reading: Readwise into your maps

Reading is how most knowledge gets in, and the failure is that highlights die in the app you read them in. The fix is a pipeline that carries them somewhere useful.

  • Capture highlights as you read (Kindle, articles, PDFs).
  • Readwise collects them all in one place and syncs them out, including into Obsidian, automatically.1
  • Then the map-refinement loop takes over: those highlights land in your vault as raw material, and you metabolise the genuinely new ones into the relevant map, linking the source back.

The point isn't to collect highlights; it's to route them to the place where they'll sharpen a map. A highlight that never leaves the reading app is a highlight you’ll never use. The pipeline (read → Readwise → Obsidian → into the map) is what makes reading compound instead of evaporate.

The anti-library and RAG

Here’s the honest problem: reading is the slowest way to get information into your head, and you will never read everything worth reading. So stop pretending you will, and build the two-part system that gets you the value anyway.

The anti-library is your collection of books you haven’t read but know you’ll want one day.2 It’s not a trophy shelf of finished books; it’s a research asset of unread ones. Mine is a Google Sheet of 1,000+ titles, each tagged by topic, so I can pull “everything I have on, say, negotiation or hypertrophy” in one filter. The tags are what turn a pile into a tool.

RAG (retrieval-augmented generation) is the supercharger on top. Point an AI at that library (or your notes, or your maps) and ask a question; it retrieves the relevant passages and answers from them, with sources, without you reading the whole book first.3 Reading is good but slow; retrieval is the supercharger. You get the specific answer you need now, and the book is still there to read in full if it earns it.

This is the toolkit and the knowledge layer fusing

Notice this is PARA and the Operating System cashing out: a tagged library (structured data) plus retrieval logic (RAG) is a tool that answers questions, not a shelf that stores books. The tactical and knowledge layers are the same idea at different scales.

Passwords: just use the native manager

The shortest section on purpose. Password management is critical and people overthink it. Just use your platform's native manager (Google Password Manager, or Apple’s Keychain): it generates strong unique passwords, syncs across your devices, autofills, and warns you about breaches, for free, with no extra app to maintain. Turn on a passkey or two-factor on the account that holds it, and you’re done. The goal here is “solved and forgotten,” not “optimised.” Don’t spend an afternoon comparing password vaults; spend ten minutes turning the built-in one on.

Part 5 Takeaways

  • The literacy layer has the highest hourly return and lowest status. Pick your two or three weakest tactics; don’t do all at once.
  • The big shift is AI as the default interface: describe the outcome instead of navigating menus, but it’s only as good as your instructions and knowledge.
  • Email: inbox zero is finally easy with scheduled AI triage and batch unsubscribing. The inbox is a queue, never a to-do list.
  • Shortcuts, command palettes, text expanders: collapse “navigate to the thing” into “name the thing.”
  • Reading: route highlights through Readwise into Obsidian and into your maps, or they evaporate.
  • Anti-library + RAG: keep a tagged library of unread books; ask it questions instead of reading everything. Reading is slow; retrieval is the supercharger.
  • Passwords: use the native manager, turn on 2FA, move on.

Your Tactical Toolkit Task List

Pick two or three

  • Connect an AI assistant to your email and set up a weekly triage + unsubscribe pass.
  • Learn the ten keyboard shortcuts for the actions you do most; set up three text expanders.
  • Turn on AI-assisted scheduling and protect one daily deep-work block on your calendar.
  • Set up the read → Readwise → Obsidian pipeline so highlights land in your vault.
  • Start (or tag) your anti-library, then point an AI at it and ask it a real question to feel the RAG difference.
  • Turn on your native password manager and 2FA. Ten minutes, then forget it.

Sources & references

Footnotes

  1. Readwise (and Readwise Reader) aggregates highlights from Kindle, articles, PDFs, and podcasts, with an official Obsidian sync that writes highlights into your vault. The principle is pipeline-agnostic: any route that moves highlights out of the reading app and into your notes works; the value is in the routing, not the specific tool. ↩

  2. The “anti-library” concept comes from Nassim Nicholas Taleb, The Black Swan (2007), via Umberto Eco’s collection of unread books treated as a research tool. Tagging by topic is the practical upgrade that makes a large anti-library retrievable. ↩

  3. Retrieval-augmented generation (RAG) couples a retrieval step (find the relevant passages in your documents) with a generation step (an LLM answers using those passages), which both grounds answers in your sources and lets you query material you haven’t read. Applied to a personal library or note vault, it turns a static collection into a question-answering tool, connecting directly to the Operating System idea. ↩

Graph View

Backlinks