Workflows
Every step-by-step workflow, explainer, and comparison DeskTheory has published, grouped by topic. Pick the topic that maps to your week.
Claude Code 54
The focused-work CLI: memory, routines, skills, hooks, and the terminal workflows CEOs run daily.
-
The email command center: five inboxes, one digest, nothing sends without you
A Claude Code skill that scans every inbox you own on a schedule, classifies each thread, drafts replies in your voice, and hands you one prioritized digest. You approve sends by number; nothing goes out without you. Every correction you make becomes a standing rule, so triage gets sharper every week.
-
When to use /goal vs. /loop in Claude Code
Claude Code has two commands that keep it running without you re-prompting, and they aren't interchangeable. /goal drives one job forward until a finish line you define is provably met. /loop re-runs a prompt on a timer to poll something. Here is the one test for which to reach for, and the mistake to avoid.
-
What is the Compound Engineering plugin for Claude Code?
A plain-English guide to the free Compound Engineering plugin for Claude Code: the plan, build, review, learn loop it adds, the 37 skills and 51 agents it ships, the /ce- slash commands that drive it, and why its compounding-memory design matters for CEOs, not just coders.
-
Claude Code in the terminal vs the desktop app: which should you use?
Claude Code runs the same agent in the terminal and in the desktop app. This breaks down what each does best: the app for visual diffs, parallel sessions, and a gentle on-ramp; the terminal for scripting, piping, and unattended automation that compounds. Plus how to run both on one machine and switch between them.
-
When to use a goal vs. a dynamic workflow in Claude Code
Two new Claude Code features that both keep Claude working on its own. A dynamic workflow goes wide: many agents on one big job at once. A goal keeps going: turn after turn until a finish line you define is provably met. Here is the test for which to reach for, and how to combine them.
-
What are dynamic workflows in Claude Code?
Dynamic workflows let Claude Code plan a big job, run tens to hundreds of agents on its pieces in parallel in one session, check the work, and hand you one finished result. It raises the ceiling from one task at a time to the whole job.
-
Memory and Projects for beginners: what they are and when to switch them on
Every chat you open starts with no memory of you, and once you understand why, the two fixes take about ten minutes to set up.
-
An AI chief-of-staff for your calendar
An AI that reviews your week the way a great chief of staff would: flagging conflicts and back-to-backs, defending your deep-work time, and briefing you on every meeting before it starts.
-
Turn a messy brain-dump into a prioritized week
A stream-of-consciousness dump of everything in your head, turned into a ranked weekly plan mapped to your real goals, with the "only you can do this" items separated from the "delegate this" ones.
-
Build your first software tool in Claude Code
Describe a tool out loud. Forty minutes later, open a working app in your browser. No code, no engineer, no budget.
-
Claude Code tips for non-technical CEOs
The keyboard shortcuts are a different page. This is how to actually work with it when you can't read a line of the code it writes.
-
Inbox zero for CEOs: triage and draft replies in the terminal
An AI pass over your inbox that sorts the last day's mail into what needs you, what's just FYI, and what's noise, then drafts replies in your voice as Gmail drafts you review and send.
-
What are connectors in Claude Code?
The plug that lets Claude reach into the tools you already use, your email, Slack, your numbers, and do the work there instead of just talking about it.
-
What is a Claude Project?
A Claude Project is a container that remembers your business context, so you stop re-pasting the same three paragraphs into every fresh chat.
-
What is gbrain?
gbrain is an open-source memory system that lets your agent remember decisions, people, and context across sessions instead of forgetting by Thursday.
-
What is MCP?
MCP is the open standard that lets AI apps connect to your CRM, calendar, and tools through one common port instead of one-off integrations.
-
What is RAG?
Retrieval-augmented generation. The trick that turns an AI that knows the internet into an AI that knows your company. Every "chat with your docs" tool you have ever touched runs on it.
-
What is the AskUserQuestion tool?
The built-in tool Claude Code uses to stop mid-task and ask you a quick multiple-choice question instead of guessing. The closest thing the terminal has to a chief of staff who checks in before…
-
What is vibe coding?
Describing software in plain English and letting an AI write it, without ever reading the code. The fastest way a non-technical CEO can get a working tool. Also the fastest way to ship something…
-
Why AskUserQuestion is one of the best features in Claude Code (and totally underrated)
Nobody puts it on the highlight reel. For a CEO who can't read the code, it might be the single most valuable thing the tool does.
-
Codex vs Claude Code
I run Claude Code every day. Codex is the coding agent your team is arguing about, and there is a good chance you already pay for it without knowing. Here is how the two differ, where each one wins,…
-
What is a coding agent?
A coding agent ships real software end to end: it writes the code, runs the tests, and opens the pull request. The tool category every CEO should know.
-
What is agentskills.io?
The open standard that makes the skills you build portable. Write a capability once, and run it on whatever tool you use next.
-
What is Codex?
OpenAI's coding agent. The same idea as Claude Code, wearing an OpenAI badge, and probably already bundled in the ChatGPT plan you pay for every month.
-
Book ideas from meetings: the source material is already being generated
If you've ever thought about writing a book, this workflow makes it tractable. Claude organizes meeting moments against your book outline. Every week.
-
Content ideas from meetings: never run out of things to write again
Claude scans your transcripts for stories, anecdotes, customer truths, and sharp observations. Output is a running ideas file you mine for LinkedIn, articles, podcasts, talks.
-
CRM enrichment from calls: every call updates your CRM for you
Every customer call auto-populates your CRM with the people, the priorities, the personal context, and the real next step. Your sales team opens a record that's already filled in.
-
Customer call to follow-up or spec: two outcomes from every call
End a customer call, run one prompt against the transcript, and get back either a draft follow-up email in your voice or a draft product spec for your team.
-
Your leadership doctrine: the principles you already use, captured
A doc that grows over time, capturing the principles, metaphors, and frameworks you actually use in meetings. The seed manuscript of how you lead, pulled from how you actually lead.
-
Make a skill in Claude Code
Build one skill. Run it Friday. Watch your weekly review write itself.
-
Meeting effectiveness review: reclaim four hours a week
Weekly review of which meetings produced decisions, which produced noise, and which could be cut.
-
Self-coaching from meetings: where am I actually avoiding hard things
Ask Claude where you've been ducking confrontation, dominating airtime, or rushing past concerns. Specific moments. Named patterns. Better alternatives. All grounded in what you actually said this…
-
Set up your CLAUDE.md file
Open a folder. Paste a template. Replace six placeholders. Test that it works. Done.
-
Terminal tricks for Claude Code
Top ten, then everything else. Print the cheat sheet when you're done.
-
Team todos from meetings: stop chasing follow-up
Action items from every meeting land in each owner's Slack DM minutes later. They have their list. You stop chasing.
-
What is a cloud VM?
A computer you rent that lives in someone else's data center. Always on. Accessible from anywhere. Where you run an agent when you do not want to keep your laptop open.
-
What is a markdown file?
A markdown file is plain text with light formatting. It is the format your AI tools use for everything from CLAUDE.md to the articles your agent drafts.
-
What is a SQLite database?
A whole database that lives in a single file on your computer. No server. No admin. No login. The thing most agent [harnesses](/articles/what-is-a-harness) use to give your agent a memory.
-
Pre-meeting brief: walk into every meeting already prepared
A brief lands on your phone before every meeting. You walk in already knowing where you left things, what's outstanding, and what to lead with.
-
What are skills in Claude Code?
The folder that turns a workflow you ran once into a capability Claude executes on command, forever.
-
What is a harness?
The model is the engine; the harness is the car. A plain-English explainer of the layer that turns a smart model into an agent that runs real work for you.
-
What is a slash command?
A prompt you save once and invoke from any Claude Code session by typing `/name`. The lightest-weight way to ship a skill.
-
Automated pipeline hygiene: stay on top of every deal without a pipeline review
A Claude Code agent reads every customer transcript from the prior day, cross-references it with your CRM, and surfaces deal moves, risk signals, and next-step gaps to Slack. You'll know where every…
-
The commitment ledger: every promise you made (or made to you) tracked
A running file that captures every commitment you made and every commitment people made to you. Refreshed weekly from your meeting transcripts. Walk into every 1:1 knowing what's outstanding on both…
-
Weekly team and investor updates, generated from your meetings
Two drafts every Friday morning. One team-facing, one investor-facing. Both pulled from the week's meeting record. You edit for tone and ship.
-
Ask your meeting history anything: the CEO's query layer
Open Claude. Point it at your meeting library. Ask any question in plain English. Get answers in thirty seconds, with citations.
-
Persistent memory across Claude Code sessions
Index your notes, decisions, and past work into a local search engine. Pull the relevant pieces into any Claude Code session with one command: `/recall`.
-
Run autonomous workflows 24/7 with Claude Code Routines
Set one recurring task to run on Anthropic's servers, with your laptop closed.
-
Granola → markdown: the foundation everyone's missing
Export every meeting to [markdown](/articles/what-is-a-markdown-file). Make it searchable in Claude. Unlock six workflows on top of it.
-
OpenCLAW vs Hermes
I run [OpenCLAW][1]. [Hermes][2] is the open-source [harness][6] most CEOs are asking me about. Here is how the two differ, what each one is good at, and the one place where Hermes pulls clearly…
-
What is a CLAUDE.md file?
CLAUDE.md is the project-memory file Claude Code reads at the start of every session, so your agent already knows your business before you type a word.
-
What is a cron?
The thing that makes your agent wake up on its own and do things for you without you having to be there to ask.
-
What is Hermes?
The open-source agent harness from Nous Research. Mobile-first. Multi-model. The main alternative to the OpenCLAW harness I run.
-
What is OpenCLAW?
OpenCLAW is the open-source harness that turns Claude into an agent that reads your inboxes, drafts in your voice, and texts you the result.
OpenCLAW & the agent harness 5
The always-on harness layer: OpenCLAW, Hermes, and the infrastructure that runs agents while you sleep.
-
How do I keep my OpenCLAW agent from posting things on the internet?
The three controls that keep an always-on OpenCLAW agent from sending or posting on its own: a human-approval gate on every outbound action, a tight allowlist of who and what it can reach, and a weekly review of what it did. Plus why the defaults leave the guardrails up to you.
-
What is Microsoft Scout and how can you use it?
Microsoft's first always-on "Autopilot" agent. It runs on the exact open [harness](/articles/what-is-a-harness) operators here already build with, [OpenCLAW](/articles/what-is-openclaw). The…
-
What is Google Antigravity?
Google's agent harness. The engine room underneath Gemini Spark, and Google's answer to the open harness layer that OpenCLAW and Hermes occupy.
-
What is Google Spark and how can you use it?
Google's 24/7 personal AI agent. The mainstream, no-terminal version of the "agent that works while you sleep" that operators here build with Claude Code and OpenCLAW.
-
What is Granola?
Granola captures a verbatim transcript and clean summary of every meeting while you stay in the conversation. The foundational layer of a CEO's AI stack.
Agent Ops 5
Running agents in production: evals, observability, orchestration, and the operating discipline around them.
-
AI agents vs AI assistants: which one for which job?
Assistant: you ask, it responds, you steer. Agent: you hand over a goal, it plans and acts on its own. The only question that matters is which one fits the job in front of you, and the honest answer…
-
A competitor-monitoring routine that pings you on real moves
A scheduled job that watches your competitors for you and alerts you only when something genuinely changes, so you stop missing the moves and stop drowning in noise.
-
Your daily executive brief, assembled before you wake up
Wake up to a one-page brief assembled while you slept: overnight items, today's calendar, your key number, and the decisions waiting on you.
-
What is a context window?
A context window is the slice of text a model can hold in mind at once. Why long chats drift, and how to keep your AI from forgetting your instructions.
-
What is an AI agent?
An AI agent figures out the steps and keeps going until the job is done. What separates agents from chatbots, explained for CEOs with a real example.
Granola & meetings 3
Turn every meeting into leverage: capture, transcript pipelines, recall, and the workflows on top.
-
Turn one customer call into a published case study draft
A structured case-study draft built from the transcript of a happy-customer call, with every metric flagged for the customer to confirm before you publish a word.
-
Turn 20 customer calls into a product-roadmap signal
A ranked, quote-backed read of what your customers actually keep asking for, extracted from twenty real call transcripts, so your roadmap is driven by evidence instead of the last loud conversation.
-
Turn interview transcripts into a structured scorecard
A consistent, evidence-backed scorecard for every candidate, generated from the interview transcript against the same rubric every time, so panel decisions stop being a clash of gut feelings.
AI for Hiring 6
Sourcing, screening, and the hiring loop, run with AI leverage.
-
A 30/60/90 onboarding plan drafted for every new hire
A role-specific first-90-days plan, drafted in minutes from the job description and your org context, so every new hire ramps fast on a real plan instead of figuring it out alone.
-
Write a job description that filters for the right operator
An outcome-driven job description that makes the right people lean in and the wrong people bounce, so your applicant pool is smaller and better before you screen a single resume.
-
Screen a stack of resumes without reading all of them
A structured summary of every applicant against your job-related criteria, so you spend your attention on the 20 candidates worth a real look instead of slogging through 280 obvious passes. The…
-
What is adverse impact?
The legal idea that a hiring practice can be illegal even when you had zero intent to discriminate, if it screens out a protected group at a meaningfully higher rate. It is the single biggest reason…
-
What is an ATS?
Applicant tracking system. The software where every job application lands, gets sorted, and moves through your hiring pipeline. If you have ever posted a role and watched 300 resumes pile up…
-
What is an HRIS?
Human resources information system. The system of record for your employees: payroll, benefits, time off, onboarding, the org chart. Everything about a person from the day they accept the offer to…
AI for Finance 6
FP&A, reporting, and the spreadsheet work, done with AI in the loop.
-
Turn your P&L into plain-English answers you can ask anytime
Upload your P&L to a finance workspace and ask it questions in plain English. Grounded answers with the math shown, no waiting on a custom analysis.
-
Board-ready financial summaries drafted from your raw numbers
The financial narrative for your board deck, drafted from your actual statements in your usual format, so you spend your time on judgment and fact-checking instead of staring at a blank page.
-
Cash-flow forecasting a CEO actually trusts
A rolling 13-week view of the cash coming in and going out, built from your own receivables and payables, that you can stress-test with a question: "what happens if our biggest customer pays 30 days…
-
Close your books faster: a CEO's month-end AI workflow
A same-day, reconciliation-ready fix list built from your accounting export, so your close stops dragging into the second week of the month.
-
Catch billing leaks and rogue SaaS spend with an AI expense review
A cancellation-and-renegotiation list built from twelve months of transactions, surfacing every duplicate tool, silent price increase, and charge nobody remembers signing up for.
-
A senior analyst, baked into Excel
Install one plugin and Excel goes from a tool you fight with to a senior analyst who builds models, reads pasted images, and rewrites entire workbooks on command.
Frontier Watch 6
What just shipped at the frontier and what it means for an operator.
-
What is Claude Mythos?
Claude Mythos is Anthropic's vulnerability-finding AI, a frontier model that hunts zero-day security holes faster than any human team. Here is what operators need: what it does, why Anthropic refuses to sell it, the Project Glasswing rollout, and what it changes about the software your company runs on.
-
What AI can and cannot do for a CEO right now
In a Harvard and BCG experiment with 758 consultants, AI made them about 40% better at some tasks and about 19% worse at others. The tasks looked identical. That gap is the whole game.
-
How AI actually works (plain English)
Under the hood of every chat tab: a pattern machine that predicts the next word, one token at a time. Beautiful fluency, no built-in truth filter, and a working memory that erases itself when the…
-
ChatGPT vs Claude vs Gemini: which should your company standardize on?
Three excellent models, one practical question: which one do you make the company default? The honest answer depends less on benchmarks than on what you already run and what you want AI to do.
-
What is a frontier model?
The handful of most capable AI models at the leading edge, the ones that cost hundreds of millions of dollars to train and reset the bar every few months. When the news says "the frontier," this is…
-
What is a large language model (LLM)?
An LLM is a prediction machine for words, not a database of facts. Knowing the difference tells a CEO exactly when to trust it and when to verify.
Getting started 25
New to AI, or only ever used ChatGPT in a browser? Start here. Nothing installed, every term defined.
-
The one-paragraph ground rule to give AI before you delegate
A single paragraph, pasted at the start of any session where you hand AI something real, that installs a pause before the session starts and cuts the corrections you'd otherwise repeat every time.
-
Using AI on your phone between meetings
Five things you can do between meetings that used to require a desk.
-
Paste a screenshot, upload a PDF: working with documents and images in AI
Stop retyping information that already exists. Three inputs, plain chat, no setup.
-
Draft and rewrite emails that still sound like you
How to use a plain chat window to draft, rewrite, and tighten emails across every situation you face, without losing the voice your relationships are built on.
-
Explaining AI to a skeptical board member or co-founder in 5 minutes
The skeptic across the table isn't wrong to be skeptical. They've watched blockchain absorb budget and big data absorb headcount, and they're not ready to do it again.
-
First-draft any document from a three-bullet brief
Collapse the blank page from 60 minutes of staring to 10 minutes of editing, for any document you regularly need.
-
The follow-up is the skill: stop accepting the first draft
The model isn't done when the first output lands. You are.
-
Get up to speed on any topic before a call
No materials, no expert on speed dial. Just a chat window and the right prompt.
-
Give it context: the one habit that multiplies every answer
Context is one of the four ingredients in every prompt. It's also the one most CEOs skip. This article goes deep on that single ingredient, shows you what it does, and hands you a reusable block you…
-
How to check AI's work: a CEO's 3-step verification habit
A lawyer was fined after most of the citations in their AI-drafted brief turned out not to exist, and they never opened a single one.
-
How to write a prompt that actually works
Most CEOs write prompts like search queries. Short, decontextualized, and hopeful. There is a better way, and it takes less than a minute to learn.
-
Prep for a meeting by pasting in the agenda
The core move is five minutes and a paste. No Granola, no terminal, no meeting history.
-
Turn rambling notes or a voice memo into clean structure
Paste the raw mess into a plain chat, name the output shape you want, and get something you can act on.
-
Spotting a confidently wrong answer before it burns you
A lawyer filed a brief packed with real-looking case citations, all invented by AI, and didn't catch a single one before it landed in court.
-
Summarize any long document or email thread in 60 seconds
One pattern, six ready-to-use lenses, and a reusable template. No terminal, no setup.
-
Show, don't tell: use examples to get the exact format you want
The fastest way to get the model to match your format is to show it one thing that already does.
-
Talk to AI with your voice: dictation and voice mode for busy CEOs
Two ways to use Claude when you cannot type. The commute, the walk, the gap between meetings: all of it is live time now.
-
When not to use AI: the work to keep human
A lawyer submitted a brief with six case citations. Every case was invented by AI. The citations looked completely real. The lawyer paid the price.
-
Desktop, browser, or phone: where should a CEO use AI?
The surface is just the window. Your account is the home.
-
Why AI gave you a bad answer, and how to fix it
You didn't break it. You gave it nothing to work with.
-
Your first 30 minutes with Claude
Sign up, lock down your privacy, build your first Project, run a real task, iterate once. Thirty minutes and you will have done something useful with it, not just played with it.
-
Free vs paid AI: what a CEO actually needs to pay for
Most CEOs are paying $20 a month and thinking they solved the AI decision, but they may be training the model on their most sensitive conversations without knowing it.
-
What is a prompt?
Everything you type into an AI chat is a prompt. But a prompt is more than a question. It's the full package you hand the model: task, context, format, and constraints. Most first prompts deliver…
-
Is your data safe in AI?
The honest answer depends on one thing almost no one checks: which plan you are on. Consumer plans and business plans treat your data completely differently, and most leaks happen because someone…
-
What is shadow AI?
Your team is already using AI for work. Just not the AI you approved, on accounts you cannot see, with data you would never have signed off on. That gap between the AI you sanctioned and the AI they…
Productivity & operating 12
The day-to-day operating workflows: drafting, delegation, policy, and keeping customer data safe.
-
The one AI habit to model so your team actually copies it
Pick one visible habit, do it this week, and narrate it. That is the whole plan.
-
Run AI office hours that actually change behavior
A standing 30-minute weekly session where someone brings a real task they're stuck on, a champion solves it live with AI, and the working prompt goes into a shared library, so adoption becomes a…
-
An AI usage policy your team will actually follow (with template)
A one-page AI policy, short enough that people actually read and remember it, that tells your team the approved tool, the data rule, and the one hard line, with a copy-paste template to start from.
-
Set up AI for your company without leaking customer data
The one-hour setup that makes the safe way to use AI also the easy way, so your team gets the tool they want and your customer data stays out of the training set.
-
Measure the ROI of AI across your company (without a data team)
A simple baseline-and-after method that turns "AI is helping, I think" into a defensible number, using a spreadsheet and a few weeks of honest measurement, no analytics team required.
-
The Monday market-and-competitor scan you run in five minutes
A standing prompt that scans your competitors and category for what actually changed in the last week and hands you a one-page brief with source links, every Monday morning.
-
Draft personalized outbound that doesn't read like a template
Per-prospect outreach grounded in one specific, true, recent fact about each person, drafted by AI in seconds and approved by a human before it sends, so your cold emails read like you actually…
-
Pressure-test a big decision with an AI red team
Write a short memo and let an AI red team run the premortem, argue against you, and surface the assumptions your big decision secretly rests on.
-
Repurpose one podcast or webinar into a week of assets
One recording, turned into a long-form article, clip scripts, a carousel, a newsletter section, and a handful of social snippets, all in your voice, from a single transcript.
-
Roll out AI to a non-technical team in 30 days
A focused 30-day rollout that picks one painful workflow, gives one team the tool and the prompts to fix it, and measures the result, instead of handing everyone a chatbot and hoping.
-
A weekly content engine: one founder voice, ten posts
A workspace loaded with your voice that turns one or two raw thoughts into a week of posts that sound like you, not a marketing bot, in about 20 minutes a week.
-
What's safe to put into AI: a CEO's data-handling guide
A traffic-light rule your whole team can remember: what's always safe to put into AI, what needs care, and what should never touch a consumer account.
Get three workflows like these every Thursday
A free weekly email. Three workflows that put you in the top 1% of CEOs. 90-second read. Every card links back to a step-by-step guide like the ones above.
Get the newsletter →