weekly-review

Weekly Review v1.0 — Cross-Area Life OS Review

Run a comprehensive weekly review across all life areas following the Integrated Life Planning OS. This is the top-level orchestration skill — it delegates to domain specialists where they exist and directly pulls data for areas without dedicated skills.

DESIGN PRINCIPLE: This skill mirrors the weekly review template in Notion (179dc01658ad81b69a2dcedb5104dbc7). The structure is: Housekeeping → Accomplishments → Reflection (per life area) → Planning → Sharing → Clean-up → Issue Log.

Life Areas

The review covers 8 life areas (consolidated from the 10-area Integrated Life Planning Template):

#	Area	What it covers	Primary data sources
1	Body	Physical health, fitness, nutrition, body composition, sleep	Oura, Garmin, Withings, Yazio, Streaks
2	Shadow	Emotional health, mental health, journaling, coaching, stress	Oura (stress, resilience), journals (manual)
3	Mind	Intellectual growth, KMS, LMS, information diet, learning	Notion, Obsidian, newsletters, Streaks
4	Spirit	Spiritual health, meditation, contemplative practices	Manual input, Streaks
5	Relationships	Partner, family, friendships, network, CRM	Clay, Google Calendar, Fireflies
6	Lifestyle	Quality of life, travel, experiences, subscriptions, admin	Email, Calendar, Notion tasks
7	Wealth	Finances, investments, taxes, company financials	HubSpot, email (invoices), Notion
8	Work	Neurons Lab (founder/CTO/BDM), team, delivery, pipeline	HubSpot, Fireflies, Slack, Notion, Gmail, Calendar

Context

• User: Alex Honchar, Co-Founder & CTO at Neurons Lab
• Location: Madrid, Spain (Europe/Madrid timezone)
• Google accounts: [email protected] (personal), [email protected] (brand/content), [email protected] (work)
• Weight goal: 85 kg (currently ~90 kg, goal is "lose")
• Step goal: 10,000/day
• Water goal: 2,000 ml/day
• Calorie base goal: ~1,972 kcal (adjusted with activity)
• Protein goal: ~96g (from Yazio, user typically exceeds this)

Reasoning Order

Before pulling data, determine the review window:

1. Default: The current ISO week (Monday to Sunday). If today is Monday, review the prior week.
2. Custom: If the user specifies a date range, use that.
3. Comparison window: Always compare against the immediately prior ISO week.

Phase 1: Parallel Data Pull

Execute ALL of the following in a single parallel tool-call batch. Use ToolSearch first to load any deferred MCP tools needed.

1A. Load Deferred Tools

Before calling MCP tools, use ToolSearch to load tool schemas for any deferred tools:

• select:mcp__garmin-connect__get_activities,mcp__garmin-connect__get_weekly_volume,mcp__garmin-connect__get_daily_overview
• select:mcp__oura__get_daily_sleep,mcp__oura__get_daily_readiness,mcp__oura__get_daily_stress,mcp__oura__get_daily_resilience,mcp__oura__get_daily_cardiovascular_age
• select:mcp__withings__withings_get_body_composition
• select:mcp__yazio__get_user_daily_summary,mcp__yazio__get_user_water_intake,mcp__yazio__get_user_goals
• select:mcp__claude_ai_Google_Calendar__gcal_list_events,mcp__claude_ai_Gmail__gmail_search_messages
• select:mcp__claude_ai_Slack__slack_search_public_and_private,mcp__claude_ai_Notion__notion-search
• select:mcp__claude_ai_Fireflies__fireflies_get_transcripts,mcp__claude_ai_HubSpot__search_crm_objects,mcp__claude_ai_HubSpot__get_user_details
• select:mcp__claude_ai_Clay__get_user_information
• select:mcp__google-workspace__manage_accounts,mcp__google-workspace__manage_email

Run all ToolSearch calls in parallel in a single batch.

1B. Pull Data (all in parallel after tools are loaded)

Body data:

• mcp__garmin-connect__get_activities — limit 20, includeSummaryOnly true
• mcp__garmin-connect__get_weekly_volume — current ISO week, includeTrends true
• mcp__garmin-connect__get_daily_overview — one call per day of the review week
• mcp__oura__get_daily_sleep — startDate/endDate = review week
• mcp__oura__get_daily_readiness — startDate/endDate = review week
• mcp__oura__get_daily_stress — startDate/endDate = review week
• mcp__oura__get_daily_resilience — startDate/endDate = review week
• mcp__oura__get_daily_cardiovascular_age — startDate/endDate = review week
• mcp__oura__get_sleep — startDate/endDate = review week (detailed sleep architecture)
• mcp__withings__withings_get_body_composition — latest
• mcp__yazio__get_user_daily_summary — one call per day of the review week
• mcp__yazio__get_user_water_intake — one call per day of the review week
• mcp__yazio__get_user_goals — current goals

Work data:

• mcp__google-workspace__manage_accounts — operation: list (discover all accounts)
• mcp__google-workspace__manage_email — operation: triage, for each of 3 accounts
• mcp__claude_ai_Google_Calendar__gcal_list_events — review week, primary calendar, maxResults 100
• mcp__claude_ai_Fireflies__fireflies_get_transcripts — fromDate/toDate = review week, limit 50, format json
• mcp__claude_ai_HubSpot__get_user_details — current user context
• mcp__claude_ai_HubSpot__search_crm_objects — objectType DEAL, open deals, sorted by closedate

Mind & Relationships data:

• mcp__claude_ai_Notion__notion-search — "weekly tasks review goals projects", page_size 25
• mcp__claude_ai_Notion__notion-search — "week [N] tasks bizdev delivery [month] [year]", with created_date_range filter
• mcp__claude_ai_Slack__slack_search_public_and_private — "after:[start] before:[end+1]", sort timestamp, limit 20, concise
• mcp__claude_ai_Clay__get_user_information — user context

Streaks (for Body + Mind):

• Run shortcuts run "Streaks Export" via Bash in this same parallel batch.
• If shortcut fails, note it and continue (unlike body-cadence-review, the weekly review does not hard-gate on Streaks — it produces a degraded review with a warning).

1C. Secondary Pulls (dependent on Phase 1B results)

After the first batch completes:

• If work calendar from Google Workspace API failed, use mcp__claude_ai_Google_Calendar__gcal_list_events with calendarId: "[email protected]"
• Fetch any Notion task databases or pages referenced in the search results that are relevant
• If Streaks succeeded, invoke mind:streaks-export-analysis

Phase 2: Analysis by Life Area

For each life area, analyze the data and produce a section. Use domain specialist skills where they exist.

2.1 Body

Delegate to body:body-cadence-review if available and the user wants the full body protocol. Otherwise, produce the Body section directly using the data pulled in Phase 1.

Structure:

• Training summary: Activities table (day, type, duration, distance, calories, avg/max HR), weekly volume totals, vs prior week trends
• Sleep: Daily table (score, duration, bedtime, wake time, deep sleep contributor, HRV avg, lowest HR), average sleep score, architecture analysis
• Readiness & Recovery: Daily scores, trend direction, cardiovascular age trend, stress summary, resilience level
• Body composition: Weight vs goal vs start weight, fat %, muscle mass, hydration
• Nutrition: Daily table (calories, protein, fat, carbs, tracking status), averages, vs adjusted goals, weekend tracking gap analysis
• Steps: Daily table with bar visualization, days meeting 10k goal, NEAT analysis
• Water: Daily tracking compliance
• Streaks habits adherence (if available)

2.2 Shadow

• Oura stress data: daily summary, stress_high vs recovery_high balance
• Resilience contributors trend (sleep_recovery, daytime_recovery, stress management)
• Flag if no journaling/coaching data sources are connected
• Note any emotional signals from meeting notes or Slack tone

2.3 Mind

• Information diet: categorize newsletters and content consumed from email
• Notion activity: tasks and pages modified during the review week
• Learning/content: any courses, books, articles referenced
• KMS/LMS activity: Obsidian vault activity (if accessible)
• Streaks cognitive habits (if available)

2.4 Spirit

• Flag as "no data sources connected" unless Streaks includes meditation/spiritual habits
• Surface indirect signals: sauna sessions, rest days, bedtime consistency

2.5 Relationships

• Network movements: New contacts from meetings (Fireflies attendees), Clay CRM updates
• Relationship maintenance: Who was interacted with this week (meetings, emails, Slack DMs)
• Upcoming social: Events, trips, networking opportunities from calendar + Notion tasks
• Action items: CRM updates needed, follow-ups, testimonials, introductions

2.6 Lifestyle

• Travel planning queue: Extract from Notion tasks anything related to trips, events, travel
• Subscriptions & admin: Flag expiring services, pending admin tasks (taxes, renewals)
• Calendar load vs free time: Ratio of meetings to unstructured time

2.7 Wealth

• Company financials: Revenue, projections, cash flow (from Fireflies meeting summaries)
• Sales pipeline: Active HubSpot deals, new deals this week, stale deals needing cleanup
• Personal finance signals: Tax filing status, investment activity (Numerai), subscription costs
• Receivables/disputes: Overdue invoices, payment issues

2.8 Work

• Meetings: Full list from Fireflies with duration, type classification (leadership/sales/delivery/hiring/ops), key action items extracted per meeting
• Business development: Deal movement, new pipeline, partner engagements, proposals
• Team management: Hiring pipeline (Ashby candidates), OKR status, team coordination
• Delivery: Client project status, risks, upsell opportunities
• Email inbox status: Per-account unread counts, key threads needing attention
• Slack activity: Active channels, key threads, decisions made
• Notion tasks: Open tasks, recently modified, overdue items

Phase 3: Planning Section

Generate forward-looking action items split into Personal and Business columns (matching the template):

Personal Planning

• Plan leisure and free time for next week
• Specific personal tasks carried forward or newly identified
• Health/body adjustments based on review findings
• Admin/lifestyle items with deadlines

Business Planning

• Upcoming meetings review — prep needed for key meetings
• Business development — priority lead and partner follow-ups
• Team management — tasks to assign, feedback to share, hiring actions
• Urgent action items from this week's meetings

Phase 4: Sharing Section

Identify shareable learnings from the week:

• Content ideas from business activities, meetings, or reading
• Internal sharing opportunities (team, Slack channels)
• Public sharing opportunities (LinkedIn, newsletter via Hypefury)
• Specific people to share insights with

Phase 5: Issue Log

Identify systematic patterns (not one-off incidents) that recurred this week or are recurring across weeks:

• Pattern description
• Evidence from this week's data
• Impact on goals
• Suggested structural fix

Output Contract

Structure the final review as:

# Weekly Review — Week [N] ([date range])

## Housekeeping: GTD Inboxes
[Table: inbox, status, unread count, key items]

## Accomplishments
### Binary (done/not done)
### Compounding (building over time)

## Life Area Reviews

### 1. Body
[Full body analysis with tables]

### 2. Shadow
[Stress, resilience, emotional health]

### 3. Mind
[Information diet, learning, KMS/LMS]

### 4. Spirit
[Spiritual practices or data gap flag]

### 5. Relationships
[Network, CRM, social calendar]

### 6. Lifestyle
[Travel, admin, quality of life]

### 7. Wealth
[Financials, pipeline, investments]

### 8. Work
[Meetings, bizdev, team, delivery]

## Planning
[Personal + Business columns with checkboxes]

## Sharing
[Learnings and distribution channels]

## Issue Log
[Systematic patterns with evidence and fixes]

File Saving

Always save the review to the Obsidian vault periodics folder:

PERIODICS_ROOT="$HOME/Library/Mobile Documents/iCloud~md~obsidian/Documents/AlexHonchar/100 Periodics"
mkdir -p "$PERIODICS_ROOT/Weekly/Week $ISO_WEEK"

Save as: $PERIODICS_ROOT/Weekly/Week $ISO_WEEK/[YYYY]-week-[N]-weekly-review.md

If the vault path does not exist, save to Desktop and inform the user.

Delegation Rules

• Body deep-dive requested: Delegate to body:body-cadence-review
• Calendar deep-dive requested: Delegate to work:gtd-calendar-review
• Email triage requested: Delegate to work:gtd-email-triage
• Slack triage requested: Delegate to work:gtd-slack-triage
• Streaks analysis: Delegate to mind:streaks-export-analysis

The weekly review produces a comprehensive but concise overview. If the user wants to go deeper into any area, delegate to the specialist skill.

Token Efficiency

This review pulls from 10+ MCP servers. To manage token budget:

• Use includeSummaryOnly: true for Garmin activities
• Use condenseEventDetails: true for Google Calendar
• Use response_format: "concise" for Slack searches
• Use max_highlight_length: 80 for Notion searches
• Use format: "json" for Fireflies (most structured)
• Limit email triage to operation: "triage" (top 20 per account), not full search
• For Yazio, only pull daily summaries (not individual consumed items)
• For Oura activity, skip if the response is too large (>100k chars) — use Garmin steps instead

Comparison Windows

Every metric should be contextualized:

• vs prior week: Direction (up/down/flat) and magnitude
• vs goal: Current vs target with gap
• vs baseline: Personal baseline where established (resting HR, sleep score, weight)

Do not compare against population averages unless the user has no personal baseline.

Tone

Thoughtful operating review. Numbers first, then interpretation, then recommendations. Direct and concise — tables over prose. Surface problems prominently. Use bold for violations and red flags. Keep each life area section to the key signals, not an exhaustive data dump. The user is a quantitative CTO — match that sensibility.

Safety

• Never create, modify, or delete calendar events, emails, Slack messages, Notion pages, or HubSpot records during the review — read-only analysis exclusively
• Do not expose full email addresses of external contacts in the report — use display names
• Do not fabricate data for life areas with no data sources — mark them explicitly as unavailable
• All times in Europe/Madrid timezone
• If an MCP server is unavailable or errors, continue with available data and note the gap