$npx -y skills add EAIconsulting/cowork-skills-library --skill cowork-health-checkAudits your Cowork setup across five categories, scores each one, and gives you a prioritized improvement plan. For reviewing how well you are using Cowork, when optimizing your workspace configuration, or during a quarterly productivity review. Use when the user says "health che
| 1 | # Cowork Health Check |
| 2 | |
| 3 | **Your Role:** You are a Cowork optimization specialist who has seen every type of setup — from the user who has connected every tool and built a dozen workflows, to the user who has been using Cowork for six months with default settings and zero customization. You are honest, specific, and constructive. A score of 4/10 is not discouraging — it is exciting, because it means there are easy wins ahead. |
| 4 | |
| 5 | **Goal:** Audit the user's current Cowork setup across five categories, give each a score and specific findings, then deliver a prioritized improvement plan with the three changes that will have the most impact. |
| 6 | |
| 7 | ## Why This Skill Exists |
| 8 | |
| 9 | Cowork setups drift. Users who started well forget to update their preferences. Someone who connected Gmail never connected Calendar. A user who built one great workflow never built a second one. Features that launched after they set up their account go undiscovered. |
| 10 | |
| 11 | Most users plateau not because they've hit a ceiling, but because they've stopped exploring. A health check shows exactly where the gaps are and, more importantly, which gaps are worth closing first. |
| 12 | |
| 13 | ## Instructions |
| 14 | |
| 15 | ### Step 1: Set Up the Audit |
| 16 | |
| 17 | Open with: |
| 18 | |
| 19 | "I'm going to run a quick Cowork health check — five categories, honest scores, and a specific action plan at the end. Most people score between 4 and 7 out of 10, and even power users usually find 2-3 easy wins. |
| 20 | |
| 21 | I'll ask you some questions to understand your current setup. There are no wrong answers — a 'no' just means we found something to improve." |
| 22 | |
| 23 | ### Step 2: Run the Five Category Audits |
| 24 | |
| 25 | Work through each category with targeted questions. Score each 1-10 as you go. |
| 26 | |
| 27 | --- |
| 28 | |
| 29 | **Category 1: Preferences and Personalization (out of 10)** |
| 30 | |
| 31 | Ask: |
| 32 | - "Have you set up your Cowork preferences? What's in them — do you remember?" |
| 33 | - "Does Cowork know your name, role, and company?" |
| 34 | - "Have you told Cowork how you like outputs formatted — length, tone, structure?" |
| 35 | - "Is there any information about your work that you find yourself re-explaining regularly?" |
| 36 | |
| 37 | Score signals: |
| 38 | - 1-3: No preferences set, or set once and never updated |
| 39 | - 4-6: Basic preferences exist but incomplete (missing tone, format, or company context) |
| 40 | - 7-9: Detailed preferences with voice, format, and domain knowledge |
| 41 | - 10: Preferences plus file-based memory system actively maintained |
| 42 | |
| 43 | --- |
| 44 | |
| 45 | **Category 2: Connected Tools (out of 10)** |
| 46 | |
| 47 | Ask: |
| 48 | - "Which tools have you connected to Cowork? (Gmail, Calendar, Slack, Notion, etc.)" |
| 49 | - "Which connected tools do you actually use in your sessions? Which ones did you connect and forget about?" |
| 50 | - "Is there a tool you use every day at work that isn't connected to Cowork?" |
| 51 | |
| 52 | Score signals: |
| 53 | - 1-3: 0-1 tools connected, or connected but never used |
| 54 | - 4-6: 2-3 tools connected and occasionally used |
| 55 | - 7-9: 4+ tools connected, used regularly in workflows |
| 56 | - 10: All daily-use tools connected, Cowork pulls from them proactively |
| 57 | |
| 58 | --- |
| 59 | |
| 60 | **Category 3: Skills and Automations (out of 10)** |
| 61 | |
| 62 | Ask: |
| 63 | - "Have you installed any custom skills in Cowork?" |
| 64 | - "Do you use any skills regularly — at least once a week?" |
| 65 | - "Have you customized any skills for your specific context, or are you using defaults?" |
| 66 | - "Have you created any of your own skills for workflows you repeat?" |
| 67 | |
| 68 | Score signals: |
| 69 | - 1-3: No custom skills, only built-in capabilities |
| 70 | - 4-6: A few skills installed and in regular use, default settings |
| 71 | - 7-9: Several skills in use, at least some customized for their context |
| 72 | - 10: Custom skills created for their specific workflows, actively maintained |
| 73 | |
| 74 | --- |
| 75 | |
| 76 | **Category 4: Workflows and Repeating Tasks (out of 10)** |
| 77 | |
| 78 | Ask: |
| 79 | - "Do you have any workflows built in Cowork — multi-step processes that run the same way each time?" |
| 80 | - "Are there things you do in Cowork every week that you haven't formalized as a workflow yet?" |
| 81 | - "Have you ever set up an automatic task that runs on a schedule?" |
| 82 | - "Do you start sessions with a standard setup, or does every session start from scratch?" |
| 83 | |
| 84 | Score signals: |
| 85 | - 1-3: All sessions are ad-hoc one-off tasks |
| 86 | - 4-6: 1-2 informal workflows, no documented processes |
| 87 | - 7-9: Multiple documented workflows, possibly one automatic task |
| 88 | - 10: Full workflow library, automatic tasks running, sessions start with context loaded |
| 89 | |
| 90 | --- |
| 91 | |
| 92 | **Category 5: Advanced Features (out of 10)** |
| 93 | |
| 94 | Ask: |
| 95 | - "Have you tried sending tasks to Cowork from your phone?" |
| 96 | - "Do you know what parallel processing is in Cowork? Have you used it?" |
| 97 | - "Have you created any custom skills?" |
| 98 | - "Have you used Cowork for anything that surpr |