Claude Cowork for repetitive work

I made an interactive survival guide for Claude Cowork. There are 12 sections covering everything from global instructions to connectors & computer use. Here are the highlights (link to the full thing below): 1) the delegation mindset - cowork is an autonomous agent you delegate to. not a chatbot you guide step-by-step. - describe the outcome, provide context and constraints, let Claude figure out the steps. think "brief a sharp colleague" not "dictate to siri." 2) your first 10 minutes - set global instructions in settings (your role, tone, preferences). - select a working folder. - connect your services (Gmail, Slack, Drive). - install the Chrome extension. 3) how claude picks its tools - cowork automatically chooses the best tool in this order: MCP connector first (fastest, most reliable), Chrome extension second (for web apps), computer use third (for native desktop apps). 4) connectors (MCP) - direct API access to 38+ services: Slack, Gmail, Google Calendar, Notion, Linear, HubSpot, Drive, Salesforce, GitHub. - click "+" > connectors > toggle on > OAuth login. 30 seconds each. 5) file access & folder instructions - drop a CLAUDE.md file in it, Claude reads it automatically every session. put your brand colors, naming conventions, team acronyms, report preferences. like onboarding a new hire once and never repeating yourself. 6) verification - check before you ship add "then review it and fix any issues" to the end of any request. - Claude creates the thing, then self-checks, catches errors, fills gaps, polishes the output. works for reports, emails, spreadsheets. 7) chrome browser integration - install "Claude in Chrome" and cowork can navigate pages, click buttons, fill forms, extract data, take screenshots all inside your browser. - talk to cowork and it uses your Chrome tabs as context. 8) computer use - Claude can directly control your desktop. Click, type, scroll, open apps, navigate Finder, Notes, Maps, whatever. You grant permission per app. 9) plugins & skills - plugins = pre-built bundles of connectors, workflows, and templates for specific use cases. install from the marketplace in one click. - skills = smaller reusable workflows you trigger from chat or create yourself with a SKILL.md file. 10) scheduled tasks & dispatch - schedule recurring tasks: daily briefings at 9am, weekly reports every friday, file cleanup on sundays. - dispatch lets you assign work from your phone: pair via QR code, fire off a task remotely, come back to your desk and it's done. 11) projects & memory - each project gets its own folder, memory, and config. Claude remembers decisions and preferences within the project. - project instructions override global settings. 12) claude vs. cowork vs. claude code - claude: questions, brainstorming, writing, feedback - cowork: autonomous execution, file automation, workflows - claude code: software development, debugging, refactoring complete interactive guide below...

I've been testing Claude Cowork across PMM workflows for months now. Battle cards. Research synthesis. Competitive monitoring. Sales enablement decks. The tool can produce ALL of it - as real files, on your computer, in the formats your stakeholders actually want. But the gap between a cold Cowork session and a properly set up one is enormous. A cold session gives you template output. The kind of thing that could have come from anywhere. You read it, you tweak it for 45 minutes, and you wonder why you bothered. 🙃 A configured session - with three context files, a clean folder structure, and global instructions - gives you output that sounds like your team produced it. The setup takes 15 minutes. Once. I wrote the full breakdown in Part 2 of my Mastering Claude for PMMs series - the setup, the workflows, the prompt formula, the gotchas, the honest limitations. (Link in the comments) And I made this cheat sheet so you can pin it and skip the trial-and-error phase 👇 Three things worth calling out: --> The mental model matters. Chat is a conversation. Cowork is a brief. You describe the end state, give it inputs, set constraints, and walk away. Come back to finished files in your folder. --> Context files are the unlock. Three .md files - who you are, how your brand sounds, how you want Claude to operate. Claude reads them every session. You stop re-explaining yourself. --> Handover files keep momentum between sessions. Claude has no memory across sessions (which will change soon enough). Write a HANDOVER.md at the end of each one and your next session picks up where you left off. Anything I'm missing?

Claude Skills turns AI into a teammate that already knows how you work. Here’s how to use it to save 30–40 minutes on recurring tasks. Most people still use AI like a search engine. Prompt → answer → context disappears → repeat. Which means every task starts from zero. You re-explain: ↳ your role ↳ your tone ↳ your formatting preferences ↳ your constraints ↳ your workflow Claude Skills removes that friction. Instead of re-briefing the model every time, you encode the instructions once  and Claude loads them automatically when the task appears again. That turns AI from a tool into something closer to operational infrastructure. Here’s the simple structure behind every Skill: Role Who Claude is in this context. Rules What to do, what to avoid, tone, formatting, standards. Trigger The phrase or task that automatically loads the Skill. Once written well, the model already understands the job before you even start prompting. A practical way to start: Build Skills for the tasks you repeat every week. Examples: Writing Skill Role: LinkedIn editor Rules: concise, structured, actionable, no fluff Trigger: “Write a LinkedIn post” Research Skill Role: industry analyst Rules: summarize trends, cite sources, highlight implications Trigger: “Research / analyze” Meeting Skill Role: chief of staff Rules: extract decisions, action items, owners, risks Trigger: “Summarize meeting” Each one eliminates repeated prompting. Setup takes about 10 minutes. But you can reuse the Skill hundreds of times. The important shift is this: The AI advantage does not come from better prompts. It comes from designing reusable thinking systems. ↳ writing frameworks ↳ research workflows ↳ analysis checklists Once encoded properly, they become persistent intelligence. That’s how individuals begin building what companies are trying to build at scale: an AI operating layer around their work. Prompting helps. But workflow design creates the real leverage. 🔁 Repost if this helped clarify how to turn AI into a system, not just a tool. ➕ Follow Gabriel Millien for practical insights on AI execution and building real leverage with AI. Image credit: Chris Donnelly

Most think Claude Cowork = Claude with folder access. That barely scratches the surface. The real power of Cowork isn’t the task you type. It’s what happens before and after you type it. Cowork actually runs as a 4-layer execution system: 1️⃣ Global Instructions → Your permanent identity Claude knows who you are before you say a word. Tone, role, format, preferences — loaded every session. 2️⃣ Folder Instructions → Project-level context Drop into any folder and Claude already knows the rules. Brand voice, output format, naming conventions — all pre-loaded. 3️⃣ Plugins → Domain expertise on demand Not just a generalist anymore. Finance, HR, legal, ops — each plugin changes what Claude knows. 4️⃣ Sub-agents → Parallel execution Complex tasks get broken into workstreams. Multiple Claude instances running concurrently. You just wait for the output. Most people open Cowork and start describing tasks. But the gap between a rough output and something near-final almost always comes down to setup. Practices that actually move the needle: → Write global instructions before session one — treat it like an onboarding doc → Add folder instructions per project — don’t re-brief Claude every time → Install role-specific plugins before you start a task → Use persistent agent threads for work that spans multiple days → Schedule recurring tasks — weekly reports, daily briefs, file processing — set once, runs forever Go deeper: 67 use cases organised by profession (actually wild) https://lnkd.in/gaq7vWRF 10 copy-paste workflows that work today https://lnkd.in/g5szdbNh Complete 2026 setup + prompting guide https://lnkd.in/grK3NfSu Official getting started docs https://lnkd.in/g_4kXFFD Bookmark this if you’re a knowledge worker building with Claude. The future of work isn’t using AI harder. It’s building a system that works without you

A weird experiment I've been trying the last few weeks is having Claude take over day-to-day maintenance of our apps. Seeing early signs of life that this might be possible. The setup is straightforward: we have a Slack channel called proj-claude-maintains-apps. In it, Claude Tag runs a bunch of daily routines across iOS, Android, Desktop, web, CLI, and Agent SDK: - Crash fuzzer: open the app in a simulator and tap around to find ways to crash it, then root cause and fix the crashes - Dup unifier: scans the codebase for similar-yet-slightly-divergent abstractions, and puts up PRs to unify them - Dead-code remover: removes statically unreachable code, and adds logging to suspected dead code to check if it's really dead and if so, remove it the next day - Abstraction police: fixes leaky abstractions - a bunch more.. Results have been surprisingly positive. Over the last few weeks, these routines have opened 388 PRs across our repos, 180 of which we merged after Claude Code Review + human review. We're now thinking about how to streamline this to make merging these kinds of mechanical changes easier. Claude generally gets these PRs right on the first shot, and if it doesn't, we ask Claude to tune its routines so it's better the next day. Sometimes it takes a few days of tuning. To try a similar workflow, ask Claude Code or Tag, or create some routines directly at https://lnkd.in/gqSqTtDF. A few of the actual prompts I used below. Has anyone experimented with similar workflows?

Claude Cowork lets you schedule your workflows. Here are 7 to have 80% of your search done by 7am: Most job seekers open their laptop every day with no plan. They scroll job boards. They apply for a bunch of jobs. Then they get overwhelmed, close the laptop, and tell themselves they'll try again tomorrow. Meanwhile, the people who are actually landing offers? They wake up to a pre-built plan. New roles scored and filtered. Tailored resumes waiting in their folder. Personalized outreach drafted and ready to send. The difference isn't effort. It's where the time is spent. That's why Claude Cowork's Scheduled Tasks are so powerful. You build the workflows. You schedule them with Cowork. Claude runs them at the same time every day (no prompt or manual trigger needed). But here's where the magic really happens: Compounding. Each workflow outputs a file. The next workflow opens that file and uses it for context. By the time the last one runs, it's built on everything before it. While every other job seeker has 7 ChatGPT convos open? You've got one pipeline that runs itself. Here's how I'd set it up: 5:15am — Workflow 1: The New Job Scout Searches for new roles posted in the last 24 hours. Scores each against your criteria. Anything under 70% gets cut. Everything else feeds the next 6 workflows. 5:30am — Workflow 2: Company Research Pulls recent news, team structure, and growth signals for every company from Workflow 1. Sets you up for better personalization in the next 5 workflows. 5:45am — Workflow 3: Resume Optimizer Takes your master resume, the JD, and your company research. Makes targeted optimizations for the ATS and the human making the callback decision. 6:00am — Workflow 4: Cover Letter Gen Uses the tailored resume, JD, and company brief to write a personalized cover letter in your voice. 6:15am — Workflow 5: Contact Identifier Maps 4 contact types at each target company: hiring managers, peers, internal recruiters, and recent hires. Names, titles, and LinkedIn URLs all ready for Workflow 6. 6:30am — Workflow 6: Connection Creator Researches each contact's recent LinkedIn activity, runs Google searches, and drafts personalized outreach. You review, add your layer of authenticity, and press send. 6:45am — Workflow 7: Admin & Organization Scans your email, updates your job search tracker, and rolls everything from all 7 workflows into one morning summary. The result? You open your laptop at 7am to a single file showing you: - Here's what's new - Here's how to apply - Here's what to send All you need to do is execute. PS - You can download full versions of all the prompts for these workflows here: AustinBelcak.com/7am 😡 Job search not working? We help people land great jobs 2x faster with a $44k raise, book a free 30 min call here 👉 https://lnkd.in/eyReiRpA

𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗵𝗮𝘁 𝗶𝘀 𝘁𝗵𝗲 𝗹𝗲𝗮𝘀𝘁 𝗶𝗻𝘁𝗲𝗿𝗲𝘀𝘁𝗶𝗻𝗴 𝗽𝗮𝗿𝘁 𝗼𝗳 𝗖𝗹𝗮𝘂𝗱𝗲. Useful? Yes. But not where the real leverage is. The real advantage is knowing which layer to pull - and when. 𝗛𝗲𝗿𝗲'𝘀 𝘁𝗵𝗲 𝗳𝘂𝗹𝗹 𝘀𝘁𝗮𝗰𝗸: ⇣ 𝟭. 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗵𝗮𝘁 → 𝗶𝗱𝗲𝗮𝘁𝗶𝗼𝗻 & 𝗰𝗹𝗮𝗿𝗶𝘁𝘆 → The conversational layer. → Use it when the work is still fuzzy. Drafts, summaries, plans, memos. It turns vague thinking into structure. You get clarity fast. You still execute the work yourself. 𝟮. 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 → 𝘀𝗵𝗶𝗽 𝗳𝗮𝘀𝘁𝗲𝗿 → The coding agent in your terminal. → Use it when the work lives in a repo. It edits across files, runs tests, debugs, iterates. Not autocomplete. A pair programmer that can actually touch the codebase. 𝟯. 𝗖𝗹𝗮𝘂𝗱𝗲 𝗣𝗿𝗼𝗷𝗲𝗰𝘁𝘀 → 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 𝗶𝘀 𝘆𝗼𝘂𝗿 𝘀𝘂𝗽𝗲𝗿𝗽𝗼𝘄𝗲𝗿 → The persistent workspace. → Use it when the work needs memory. Documents, instructions, context — loaded once, available always. No re-explaining. No starting from zero. This is where Claude becomes useful for serious client work. 𝟰. 𝗖𝗹𝗮𝘂𝗱𝗲 𝗦𝗸𝗶𝗹𝗹𝘀 → 𝘀𝘁𝗮𝗻𝗱𝗮𝗿𝗱𝗶𝘇𝗲. 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗲. 𝘀𝗰𝗮𝗹𝗲. → The repeatability layer. → Use it when the same task happens again and again. Skills turn your way of working into reusable instructions. The value isn't one good output. It's consistent output across people, teams, and weeks. 𝟱. 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝘄𝗼𝗿𝗸 → 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗲 𝘁𝗵𝗲 𝗯𝗼𝗿𝗶𝗻𝗴 𝘀𝘁𝘂𝗳𝗳 → The desktop automation layer. → Use it when the work isn't thinking, it's coordination. Organizing files, extracting PDFs, moving data between apps. The manual glue work most teams still waste hours on. — → Need clarity? Chat → Need to build? Code → Need consistency? Skills → Need continuity? Projects → Need execution across files? Cowork — ⇣ 𝗘𝘃𝗲𝗿𝘆 𝘄𝗲𝗲𝗸 𝗜 𝗰𝘂𝗿𝗮𝘁𝗲 𝘁𝗵𝗲 𝗯𝗲𝘀𝘁 𝗔𝗜 𝗱𝗿𝗼𝗽𝘀 - 𝗹𝗲𝗰𝘁𝘂𝗿𝗲𝘀, 𝗴𝘂𝗶𝗱𝗲𝘀, 𝗵𝗮𝗻𝗱𝘀-𝗼𝗻 𝘀𝘁𝘂𝗳𝗳 - 𝗶𝗻 𝗺𝘆 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿 𝗛𝘂𝗺𝗮𝗻 𝗶𝗻 𝘁𝗵𝗲 𝗟𝗼𝗼𝗽 [𝗙𝗿𝗲𝗲]: https://lnkd.in/dbf74Y9E

Google, Meta, and Amazon now list AI fluency as a PM job requirement. Claude Code went 0 → market leader in 8 months. You don't need to write code. Here's how to start — 7 steps, 10 minutes: 1. Download VS Code. Free. Mac, Windows, Linux: code.visualstudio.com Just an app — like Slack or Notion. 2 min. 2. Install the extension. Ctrl+Shift+X → search "Claude Code" → Install. Disable GitHub Copilot Chat if it's active. 3. Open your Cowork folder. File → Open Folder → pick the same folder you use with Cowork. Your CLAUDE.md and connectors carry over. You don't start from zero. Most people miss this. 4. Set Model & Effort. /model for Opus 4.6. /effort for thinking depth. The VS Code extension gives you UI buttons for both. 5. Your connectors follow you. Gmail, Slack, Notion — tied to your account. They work in Code automatically — no re-setup. Zero config. 6. Learn Plan Mode. Press Shift+Tab before any request. Claude proposes a plan, waits for OK before changing anything. You approve before anything happens. 7. Let Auto Memory work. Claude saves your patterns to MEMORY.md per workspace. Persists across sessions automatically. Use /memory to review or edit what Claude remembers. --- What I actually use it for: → Market monitoring — Chrome MCP watches Reddit, X, and competitor feeds for emerging signals. Runs while I sleep. → Hypothesis tracking — Claude maintains my hypotheses, logs new evidence per post, and flags when something graduates from "hunch" to "pattern." → Competitive analysis — reads 30+ posts, extracts hooks, structures, engagement data. One session, not one week. → Visual assets — generates infographics from my data and notes. I describe the layout, Claude builds it. → Building automations — "create a workflow that drafts Slack replies." → Writing — Claude knows my voice, values, and principles. It writes with me, not instead of me. None of this requires writing code. --- Commands worth knowing: /model, /effort → Switch model & thinking /context → Token usage (watch at 50%) /memory → Manage auto memory /loop → Recurring tasks (3-day) /init → Create CLAUDE.md /compact → Compress conversation --- Pro tips most guides skip: → Pipedream MCP: One connector → 1,000+ APIs. Gmail, Stripe, Jira. Free. → Remote Control: Run claude remote-control → continue from mobile. → Cascading CLAUDE.md: Root + subfolder overrides. Like CSS inheritance. → Combine Cowork + Code: Cowork /schedule for daily cron. Code /loop for in-session. Same connectors. Same CLAUDE.md. --- The setup takes 10 minutes. The advantage compounds from day one. Want to learn how to build AI products as a PM? The AI PM Certification (#1 Maven) covers it hands-on: 👉 https://bit.ly/aipmcohort You may also benefit from Claude Code for PMs: https://lnkd.in/dtsCYeVv From Paweł Huryn.

You can now give Claude memory with projects. No more starting from zero every session. Every Cowork session used to begin the same way. You’d have to re-upload files, re-explain context, and re-state what you’d already covered. Projects change this by bundling four things into one workspace: → Instructions: rules Claude follows for every task in scope → Context: reference docs, transcripts, and folders Claude reads from → Scheduled tasks: recurring work that runs automatically in the background → Memory: scoped to the project, so nothing leaks across workstreams The simple test for Skill vs Project: "Does this work need memory of what came before?" If no, it's a Skill (a recipe for repeatable tasks). If yes, it's a Project (a workspace for ongoing work). Skills are the tools in the workshop. Projects are the workshop itself. 5 steps to set one up: 1. Create a project (start fresh, import from Chat, or use an existing folder) 2. Add instructions defining the goal, terminology, and rules 3. Add context like reference docs, transcripts, and PDFs 4. Run your work inside the project 5. Stack it with Skills for repeatable formatting This is the layer that finally makes Claude feel like a real partner for thought. I broke down how to set up Claude Projects in full detail here: https://lnkd.in/e8dz8pRk

Your company doesn’t care if you “know AI.” They care how many Claude tokens you can turn into outcomes. The Claude ecosystem has 5 products each designed for a specific workflow. Here's when to use which: 1. Claude Chat AI for thinking, writing, research. Role: Thinking Partner Best for: • Ideas, writing, explanations, research • Deep reasoning & long-form responses Skip for: • Files, automation, system actions Example: • Writing a blog post about AI trends • Claude Chat researches, outlines, drafts, and refines arguments 2. Claude Cowork Desktop automation agent. Role: Workflow Automator Best for: • File tasks, batch work, automation • Cross-app task execution Skip for: • Reasoning, creative thinking Example: • Extract data from 50 PDFs, organize in spreadsheet, send summary emails • Claude Cowork automates the entire workflow 3. Claude Code Autonomous coding agent. Role: Engineering Agent Best for: • Debug, build, refactor, test • Multi-file code understanding Skip for: • Ideation, non-tech tasks Example: • Building a feature that touches 10 files • Claude Code analyzes codebase, writes code, updates tests, refactors 4. Claude Skills Reusable instruction sets. Role: Capability Extender Best for: • Templates, repeat tasks • Consistent repeated workflows Skip for: • One-time or exploratory tasks Example: • Create LinkedIn posts every week with same structure • Build a Claude Skill with formatting rules every post follows same style 5. Claude Projects Persistent workspace. Role: Context Keeper Best for: • Ongoing projects • Long-term memory & context Skip for: • Quick one-off tasks Example: • Managing a product launch with hundreds of documents • Claude Projects remembers everything context persists across sessions Each product handles a different stage. Chat thinks. Skills standardize. Projects persist. Code builds. Cowork automates. Most teams use Claude Chat for everything and hit limits. The teams winning with Claude use the right product for each workflow. Which Claude product have you been using lately? Reshare with your network to make them "AI ready" and save to refer these usecases later. #Claude #AgenticAI #AIAgents