Agent skill

Setup Cowork

by asgeirtj in asgeirtj/system_prompts_leaks

Guided Cowork setup — install a matching plugin, try a skill, connect tools.

CC0-1.0Auto-check passed

Install Setup Cowork

skills CLI
$ npx skills add asgeirtj/system_prompts_leaks --skill setup-cowork -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install asgeirtj/system_prompts_leaks setup-cowork --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/asgeirtj/system_prompts_leaks.git skills-src && mkdir -p .claude/skills && cp -r skills-src/Anthropic/claude-cowork/setup-cowork .claude/skills/setup-cowork && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
setup-cowork
GitHub stars
69k
Token cost
~3k tokens
SKILL.md length
1,979 words
Files
1
Skills in repo
128
Repo updated
First seen
Licence
CC0-1.0

At a glance

Guided Cowork setup — install a matching plugin, try a skill, connect tools.

  • Works in 7 steps: Checklist → Role → Suggest plugins → …
  • : set up cowork
  • SKILL.md covers Step 0 — Checklist, Step 1 — Role, Step 2 — Suggest plugins and Step 3 — Connectors, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Setup Cowork is an agent skill from asgeirtj/system_prompts_leaks. Guided Cowork setup — install a matching plugin, try a skill, connect tools. Use when: set up cowork, setup cowork, get started with cowork, cowork onboarding, configure cowork, personalize cowork.

Its SKILL.md is about 3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Documented system prompts from Anthropic - Claude Fable 5.1, Opus 5.5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-6-Astra, Codex. Google - Gemini 3.8 Flash, 3.1 Pro… The licence is CC0-1.0.

When your agent uses it

  • : set up cowork
  • Get started with cowork
  • Cowork onboarding
  • Configure cowork

Example prompts

  • “/setup-cowork”

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. Checklist
  2. Role
  3. Suggest plugins
  4. Connectors
  5. Try a skill
  6. Writing voice
  7. Wrap

What it can do on your machine

Read from SKILL.md and the folder at commit 60d44cc. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Setup Cowork loads about 3k tokens when it runs. Until then it costs about 53 tokens; SKILL.md has 1,979 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~53
When it runs · the whole SKILL.md, loaded when a task matches
~3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from asgeirtj/system_prompts_leaks at commit 60d44cc, republished under its CC0-1.0 licence (© asgeirtj). 1,979 words, ~2,989 tokens.

Download SKILL.mdSave it as .claude/skills/setup-cowork/SKILL.md (or your agent's skills folder).
name
setup-cowork
description
Guided Cowork setup — install a matching plugin, try a skill, connect tools. Use when: set up cowork, setup cowork, get started with cowork, cowork onboarding, configure cowork, personalize cowork.

Setup Cowork

Help the user get Cowork configured for their work. Six steps — role, plugins, connectors, try a skill, writing voice, wrap.

Step 0 — Checklist

Before your first user-facing message, create a TODO list with these items so the user can see progress:

  1. Figure out role
  2. Suggest plugins
  3. Suggest connectors
  4. Try a skill
  5. Set up writing voice
  6. Wrap up

Mark each one complete as you finish it. Keep it to these six — don't add sub-items.

Step 1 — Role

Your initial message should frame what Cowork is: it autonomously handles tasks like reading your email, searching your docs, drafting reports, etc. Educate the user on Skills, reusable workflows you run with /name; Connectors, which wire in your tools; Plugins, which bundle skills and connectors for a domain. Two or three sentences. Hit the beats: multi-step and autonomous, uses your real tools, skills/plugins/connectors defined.

Next, ask the user for their role. Something like: "Let's get you set up — takes a few minutes. What kind of work do you do?" Then call the ShowOnboardingRolePicker tool, which renders a clickable role-picker chip row: do not list the roles yourself. The tool result is their answer — {"role": ...} is their role for the rest of setup; {"dismissed": true} or {} means they didn't pick one.

If the ShowOnboardingRolePicker tool is not available in this session, ask in plain text instead and offer these options as a short list they can reply to (they can also answer in their own words):

  • Product management
  • Engineering
  • Human resources
  • Finance
  • Marketing
  • Sales
  • Operations
  • Data science
  • Design
  • Scientist
  • Legal
  • Student
  • Founder
  • Healthcare

In the plain-text case, end your turn after asking. Their reply — one of the options or a free-form answer — is their role for the rest of setup.

Step 2 — Suggest plugins

The role picker tool result will contain their selection. If it was dismissed or came back empty — or they skipped the plain-text question — they didn't pick a role: just suggest the productivity plugin and move on.

Always check for already-installed plugins before doing anything else — this is not optional. Call ListPlugins without any intro text — do not write "Looks like you already have…" before you know the result. The tool renders the installed plugins as a widget on its own; let it speak for itself. After it returns, react to what actually came back: if plugins appeared, acknowledge them below the widget ("Those are already on your account — here's what else fits your role."); if it's empty, just say "No plugins yet — let's fix that." Never write text that presumes a non-empty result before the tool runs. Do not pass installed plugins to SuggestPluginInstall afterward or you'll show them twice. Admin-provisioned plugins will appear in this list automatically; never skip the call. Then, regardless of what's installed, still recommend new role-matched plugins below in a separate widget.

Search the plugin marketplace for their role with SearchPlugins. Exclude anything already installed — the installed-plugins widget above already covers those, so the recommendations widget must only contain plugins the user does not yet have. Never show the same plugin in both widgets. Organization plugins always come first. If the user's org has published its own plugins, those are the recommendation — they're built for this company's actual tools, data, and workflows, and someone internal decided they matter. An org-built plugin that's even loosely relevant to the role outranks any generic marketplace plugin, full stop. Lead with org plugins, and only reach for generic ones to fill empty slots when the org catalog has nothing close. Never bury an org plugin under a generic one.

Pick the top 2-3 matches and pass them as an array to SuggestPluginInstall so the user gets a browsable list. If only one is a strong fit, passing one is fine. If the search comes up empty, fall back to the productivity plugin. If every good match is already installed, skip the recommendations widget entirely and just say "You've already got the best plugin for [role] — let's move on to connectors."

Above the widget, introduce it in one line: "Here are plugins built for [role] work — each one adds a set of skills you can run with /." The card shows Add or Manage depending on whether each plugin is already installed — don't describe the button. Below the widget, reinforce what they're for and tie it to the next step: "Installing one drops its skills straight into your / menu so you can run them anytime. Once you've picked one, want me to pull up the connectors it uses so those skills have your real data behind them?" — phrased so it works whether they're installing fresh or already have it. End your turn.

Step 3 — Connectors

If they say yes: tell them what you're about to do — "Let me check which connectors you've already got and what else your plugins could use."

Cover every plugin in play — everything already installed plus anything the user just added. Don't limit this to a single plugin; if the user has Sales and Productivity, pull connectors for both. Search SearchMcpRegistry per plugin domain, using the plugin's name and the user's role as queries, until every plugin in play has connector results — the results carry each connector's directoryUuid and whether it's already installed. Don't drop any relevant hit to prose; every connector those searches surface for their plugins should end up in the widget.

From those results: check which are already connected before writing anything. Only if at least one is connected, call ListConnectors with those names as keywords — and do not write "You're already connected to these:" above it; let the widget show it. If none are connected, skip ListConnectors entirely. Then call SuggestConnectors with all the still-unconnected UUIDs — the full set the searches surfaced, not just the top match. Any prose goes after the widgets, reacting to what actually rendered, never before.

Below the suggestions, explain what they're looking at before moving on: "Click any of these to connect it — once wired up, skills can pull your real data from it. Want me to list some skills you can try?" End your turn.

Step 4 — Try a skill

If they say yes, call ListSkills with the plugin's name and their role as keywords so they get clickable skill cards; if the filter comes back empty, call it again with no keywords. Introduce the card in one line so it doesn't land cold: "Here's what [Plugin] adds — click any of these to run it now." End your turn. That card is keyword-filtered — when a later step needs to know everything on the user's account (Step 5 does), the answer comes from a keywordless ListSkills call or your system context's skills list, never from this filtered card.

When they click one (you'll see a /name message), help them with it. Keep it brief; you're still inside setup. When it finishes, bring it back: "Nice — that's how skills work."

If they wave it off at either point, that's fine — go to Step 5.

Show full SKILL.md (817 more words)Show less

Step 5 — Writing voice

Everything so far taught Cowork about the user's tools. This step teaches it about the user. This matters because so much of what Cowork produces is prose the user will send under their own name.

First, settle which opener you're writing — the account's full skills list decides. Check the skills in your system context, or call ListSkills with no keywords; the plugin-filtered card from Step 4 covered one plugin and can't answer this. If my-writing-style is there (the saved profile — not setup-writing-style, the flow that creates it) — or the user says they've already set one up — your whole message is one line ("You've already got a voice profile, so anything I draft for you will use it") and you go to Step 6. Only if it's absent do you offer setup. Re-running the flow on someone who's already done it wastes their time and risks overwriting a profile they've tuned. If they want to update or redo it, that counts as a yes — invoke the skill the same way.

If the user says they already have one, that settles it — a recently saved profile may not show in your skills list yet, so their word beats the list. Never tell a user they don't have a profile on the strength of a widget result; the widgets in this flow are plugin-filtered, and silence from one means nothing. Skipping a redundant offer costs a sentence; overwriting a tuned profile costs the user their work.

If setup-writing-style itself isn't available in this session, skip the offer entirely: mark this TODO done and go to Step 6 — the wrap's closing clause covers it.

Otherwise, offer it. Make the case in two or three sentences of prose — these are the beats to hit, not a list to reproduce — then ask. Don't just launch into it:

  • What it does: reads writing they've already sent, learns how they write, and saves it so future drafts sound like them instead of like Claude.
  • What it costs: about two minutes.
  • What it protects: only writing they authored, and nothing saves without their review. (One clause — the skill itself walks through consent in detail once they say yes.)

Phrase the ask so passing is obviously fine — "Want to do that now, or skip it?" A user who feels cornered into a two-minute detour at the end of setup will just abandon the whole thing.

If they say yes: invoke the setup-writing-style skill (via the Skill tool — don't improvise its flow from memory) and let it run end to end. Don't paraphrase its steps, re-explain consent, or interleave your own commentary — it opens with its own framing, and a second voice narrating over it is confusing. Cowork setup is paused, not over. The voice flow counts as finished when one of three things happens: the save tool reports success; the user confirms the profile is saved (when saving happens via a Save skill button, you can't see the click and the new skill won't appear in your skills list until their next session — the flow already has you ask them to click it, so their answer is your signal; don't ask twice); or they ask to skip or move on to something else. Only then mark this TODO done and move to Step 6 — invoking the skill starts this step; it doesn't complete it.

If they say no or defer: mark the TODO done and tell them they can always create their voice profile later by simply asking — e.g. "No problem. Whenever you want drafts to sound like you, just ask me to learn your writing voice." Then Step 6. Don't sell it twice.

Step 6 — Wrap

Close short: "You're set. Start a new task from the sidebar anytime, or type / to see your skills."

If they don't have a voice profile by the wrap, add one clause and no more: "…and whenever you want drafts to sound like you, just ask me to learn your writing voice."

Ground rules

  • One step at a time.
  • Skips are fine. If they pass on a step, mark its TODO done and move on.
  • Keep each message short. Two or three sentences plus the widget, not a wall.
  • Never write text that presumes a tool result before the tool runs. Don't say "you already have…" or "you're connected to…" above a widget — call the tool first, then react to what came back below it. The widget shows the data; your sentence reacts to it.
  • The user trying a skill mid-flow is expected. Help with it, then return to where you left off. Don't let a skill invocation end the setup. This applies to Step 5 too: setup-writing-style is a long flow, and when it ends — however it ends — the user still needs the Step 6 wrap.
  • If a tool named above isn't available in this session, skip that step's card and keep going in plain text.

© asgeirtj, CC0-1.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in Anthropic/claude-cowork/setup-cowork of asgeirtj/system_prompts_leaks.

Open the folder on GitHubat commit 60d44cc

Compare with similar skills

Setup Cowork next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Setup Cowork compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Setup Cowork this skillasgeirtj/system_prompts_leaks69k—~3kAutomated safety check: PassCC0-1.0
ConnectComposioHQ/awesome-claude-skills77k3 repos~987Automated safety check: PassNone
Node Connectopenclaw/openclaw392k—~1.6kAutomated safety check: PassMIT
Claude Academy Guideanthropics/skills180k3 repos~1.9kAutomated safety check: PassApache-2.0
Write Guidevercel/next.js143k—~1.6kAutomated safety check: PassMIT
Design Guidepaperclipai/paperclip100k1 repos~3.1kAutomated safety check: PassMIT

Similar skills

  • Connect

    ComposioHQ/awesome-claude-skills

    Connect Claude to any app. An agent skill from ComposioHQ/awesome-claude-skills.

    77k GitHub starsUsed in 3 repos~987 tokens
    Productivity & AutomationAuto-check passed
  • Node Connect

    openclaw/openclaw

    Diagnose OpenClaw Control UI browser and native Android, iOS, or macOS node connection failures across route, auth, pairing, QR/setup-code, and reconnect states.

    392k GitHub stars~1.6k tokensUpdated today
    MobileAuto-check passed
  • Claude Academy Guide

    anthropics/skills

    Official

    Adds a matching Claude Academy course, tutorial or use case to the end of an answer about learning to use Claude, but only when the match is strong.

    180k GitHub starsUsed in 3 repos~1.9k tokens
    EducationAuto-check passed
  • Write Guide

    vercel/next.js

    Official

    Generates technical guides that teach real-world use cases through progressive examples.

    143k GitHub stars~1.6k tokensUpdated today
    DevelopmentAuto-check passed
  • Design Guide

    paperclipai/paperclip

    Paperclip UI design system guide for building consistent, reusable frontend components.

    100k GitHub starsUsed in 1 repo~3.1k tokens
    Frontend & DesignAuto-check passed
  • Write Guides

    remix-run/remix

    Write, rewrite, or audit Remix guide chapters in the voice of docs/guides/app/actions/docs/chapters/01-start-here.md.

    33k GitHub stars~2.6k tokensUpdated yesterday
    Writing & ContentAuto-check passed

More from asgeirtj/system_prompts_leaks

All 125 skills in this repo
  • Fleet Manager for Agent Sessions

    asgeirtj/system_prompts_leaks

    Shows one digest of coding-agent sessions across your connected machines and lets you open, read, steer, approve, stop and close them, over Herdr, tmux or MSP.

    69k GitHub stars~2.5k tokensUpdated today
    Auto-check passed
  • Muse Code Product Doctor

    asgeirtj/system_prompts_leaks

    Diagnoses a Muse Code installation's own failures from binary and session evidence, instead of treating the report as an ordinary repository bug.

    69k GitHub stars~3.5k tokensUpdated today
    Auto-check passed
  • DOCX

    asgeirtj/system_prompts_leaks

    A skill your agent uses whenever the user wants to create, read, edit, or manipulate Word documents (.docx) or Word templates (.dotx).

    69k GitHub stars~1.9k tokensUpdated today
    Auto-check passed
  • Agents Project Coordinator

    asgeirtj/system_prompts_leaks

    Runs a goal as a project in which the agent coordinates separate agent threads, judging when to split the work, and interviews you first when nothing can be verified.

    69k GitHub stars~2.9k tokensUpdated today
    Auto-check passed
  • Muse Plugin Creator

    asgeirtj/system_prompts_leaks

    Creates and validates a new native Muse plugin package in the current workspace, limited to five capability families, and leaves installation to you.

    69k GitHub stars~1.8k tokensUpdated today
    Auto-check passed
  • Deep Research

    asgeirtj/system_prompts_leaks

    A skill your agent uses when the user's prompt requires (1) researching a topic across multiple sources, comparing options or alternatives, analyzing trends or history, understanding markets or…

    69k GitHub stars~3.3k tokensUpdated today
    Auto-check passed

Questions about Setup Cowork

What does Setup Cowork do?

Guided Cowork setup — install a matching plugin, try a skill, connect tools. Setup Cowork is an agent skill from asgeirtj/system_prompts_leaks. Guided Cowork setup — install a matching plugin, try a skill, connect tools.

When should I use Setup Cowork?

Setup Cowork fits situations like: : set up cowork; get started with cowork; cowork onboarding; configure cowork.

How do I install Setup Cowork in Claude Code?

Run `npx skills add asgeirtj/system_prompts_leaks --skill setup-cowork -a claude-code`. Or copy the skill folder (Anthropic/claude-cowork/setup-cowork in asgeirtj/system_prompts_leaks) into .claude/skills/setup-cowork in your project. Claude Code loads it when a task matches its description.

How do I install Setup Cowork in Codex?

Run `npx skills add asgeirtj/system_prompts_leaks --skill setup-cowork -a codex`. Or copy the skill folder (Anthropic/claude-cowork/setup-cowork in asgeirtj/system_prompts_leaks) into .agents/skills/setup-cowork in your project. Codex loads it when a task matches its description.

Can I use Setup Cowork in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add asgeirtj/system_prompts_leaks --skill setup-cowork -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/setup-cowork, .gemini/skills/setup-cowork, .github/skills/setup-cowork and .opencode/skills/setup-cowork in your project.

What does Setup Cowork need to run?

SKILL.md names no scripts, command-line tools or credentials: Setup Cowork is instructions for the agent only.

Does Setup Cowork access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Setup Cowork safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Setup Cowork use?

Setup Cowork is published under the CC0-1.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Setup Cowork use?

About 3k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Setup Cowork?

Skills that share tags, products or a category with Setup Cowork: Connect (ComposioHQ/awesome-claude-skills, 77k stars), Node Connect (openclaw/openclaw, 392k stars), Claude Academy Guide (anthropics/skills, 180k stars) and Write Guide (vercel/next.js, 143k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Setup Cowork?

asgeirtj (a GitHub user) maintains it in asgeirtj/system_prompts_leaks, which has 69,280 GitHub stars. The repository holds 128 skills in this directory. The repository was last updated on October 10, 2026.

Source: asgeirtj/system_prompts_leaks on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.