Weekly Review

Interactive weekly reflection with Toggl integration and historical trend analysis.

Run on Sundays - Reviews the previous week (Sunday through Saturday).

Arguments

Required (first argument): Path to JSONL data file

•Example: /weekly-review ~/app/ignore/weekly-reviews.jsonl

The file will be created if it doesn't exist. Data is appended (one JSON object per line).

Quick Reference

•Trends: Show 8-week historical patterns with bar charts
•Toggl: Fetch and aggregate time entries by project/activity
•Context: Present pre-filled data, prompt for narration
•Narrate: User speaks freely, Claude tracks and maps to structure
•Structure: Present filled form, highlight gaps
•Challenge: Red team pushes back on patterns and blind spots
•Save: Append to JSONL file

Critical Constraints

These rules take priority over other instructions:

•
Don't condense or editorialize user prose - When organizing user's words into sections, preserve their full text (though removing filler words is good). Don't summarize "I'm still not prioritizing PMing monitorability enough. I'm not attending stand-ups yet" into "Not prioritizing well". The detail enables future red-teaming.
•
Preserve full detail for future analysis - Specific names, events, dates, and context are the value. Future red-teaming depends on being able to reference "Beth" or "Tuesday's planning meeting" - generic summaries lose this.
•
Confirm before phase transitions - Always confirm with the user before moving from narration (Phase 4) to structured presentation (Phase 5). Don't assume "done" from ambiguous signals.

Argument Parsing

The JSONL path is passed as the skill's args parameter:

code

/weekly-review ~/path/to/weekly-reviews.jsonl

Handling:

•If args is empty or missing, prompt the user: "Please provide a path to your weekly reviews JSONL file"
•Expand ~ to the user's home directory (use $HOME or equivalent)
•Create the file if it doesn't exist (first review will have no historical trends)

Phase 1: Trend Analysis

•
Load historical data from the provided JSONL path
- •If file doesn't exist, start fresh (no historical trends to show)
- •Parse last 8 weeks of entries
•
Generate trend visualizations using plotext via uvx:

bash

uvx --with plotext python -c "
import plotext as plt

weeks = ['Nov23', 'Nov30', 'Dec7', 'Dec14', 'Dec21', 'Dec29', 'Jan4', 'Jan11']
development = [30, 30, 40, 20, 30, 50, 40, 45]

plt.bar(weeks, development)
plt.title('Development ↗ up')
plt.ylim(0, 60)
plt.plotsize(40, 12)
plt.theme('clear')
print(plt.build())
" | sed 's/\x1b\[[0-9;]*m//g'

Output:

code

             Development ↗ up           
  ┌────────────────────────────────────┐
60┤                                    │
50┤                      █████         │
40┤         █████        ██████████████│
30┤██████████████    ██████████████████│
  │██████████████    ██████████████████│
20┤████████████████████████████████████│
10┤████████████████████████████████████│
 0┤████████████████████████████████████│
  └──┬────────┬────────┬───┬────────┬──┘
   Nov23    Dec7   Dec21 Dec29   Jan11

For each major activity/rating, generate a separate chart. Show 3-4 charts that are most relevant:

•Activities with biggest changes (up or down trends)
•Any ratings that hit unusually low values
•The user's top time-consuming activities

Annotate anomalies in the title or after the chart:

•"Mental: 3 ← lowest in 8 weeks"
•"Development ↗ trending up (avg 35%, last week 45%)"

•
Detect patterns:
- •Recurring themes in bottlenecks/mistakes (look for repeated words/phrases)
- •Correlation between low ratings and specific activities
- •Priority tracking: did last week's priorities appear in this week's successes?
- •Surface anomalies: "You've mentioned sleep issues in 3 of the last 5 weeks"
•
Present trends: "Here's how your past 8 weeks looked. Keep this in mind as you reflect."

Phase 2: Toggl Data Collection

•
Determine user's timezone:
- •Check system timezone or ask user if unclear
- •Default assumption: US Pacific (America/Los_Angeles)
•
Calculate date range in user's local timezone:
- •End: Yesterday (Saturday) at 23:59:59 local
- •Start: Sunday of the previous week at 00:00:00 local
- •Example: If today is Sunday Jan 18 Pacific, range is Jan 11 00:00 - Jan 17 23:59 Pacific
•
Convert to UTC for Toggl API query:
- •Toggl stores times in UTC, so expand the query range to capture all local-time entries
- •For Pacific time: query from start_date (local Sunday) to end_date + 1 day (UTC)
- •Example: Jan 11-17 Pacific → query Toggl for Jan 11-18 UTC
- •This ensures entries logged late Saturday Pacific (which is already Sunday UTC) are included
•
Fetch time entries using toggl_get_time_entries with the expanded UTC range
•
Filter results to local timezone window:
- •Convert each entry's start time to user's local timezone
- •Exclude entries outside the local date range (e.g., entries on Jan 19 UTC that are actually Jan 18 Pacific stay; entries that are Jan 19 Pacific get excluded)
•
Aggregate by project and tag to compute actual time percentages
•
Map to categories:

Activities (map from Toggl tags):

•Admin, Code Review, Data, Design, Development, Events
•Meetings, Messages, Ops, Planning, Reading, Research
•Training, Writing, Other

Projects (map from Toggl projects):

•3PRA, Capabilities, Data Pipeline, Hiring, Human Uplift
•Infra, Inspect Action, Managing, Middleman, Monitorability
•Org, Tasks and Agents, Vivaria, Other

Use fuzzy matching for project/tag names (case-insensitive, partial match).

Phase 3: Present Context

•Show computed time allocations as pre-filled tables:

code

Activities (from Toggl):
  Development: 35%
  Meetings: 20%
  Code Review: 15%
  ...

•
Display question sections for reference:
- •Time allocation (activities and projects)
- •Goal tracking checkboxes
- •Reflection questions (successes, mistakes, bottlenecks)
- •Self-assessment ratings (1-7 scale)
- •Next week priorities
•
Prompt: "Toggl says you spent X% on Development, Y% on Meetings. Does that feel right? Start narrating your thoughts - I'll track them."

Phase 4: Free-Form Narration

Let the user speak/type freely. Track and map to form sections:

•Time corrections: "Actually spent more time on Ops than Toggl shows" → adjust activities
•Goal tracking: "I paired with Thomas on Tuesday" → mark checkbox
•Reflections: "The deploy went smoothly" → successes; "Should have communicated earlier" → mistakes
•Ratings: "Productivity was maybe a 5" → track rating
•Priorities: "Next week I need to focus on the viewer" → next_week.priorities

Brief acknowledgments: "Got it - tracking as a bottleneck" (keep responses minimal)

CRITICAL: Don't condense or editorialize. When organizing user's words into sections:

•Reorganize into the appropriate fields, but preserve the user's full prose
•Don't summarize "I'm still not prioritizing PMing monitorability enough. I'm not attending stand-ups yet, I don't have daily syncs with Lawrence" into "Not prioritizing well"
•The detail is the value - future red-teaming depends on specific names, events, and context
•If the user gives bullet points, keep them as bullet points
•If the user gives long prose, keep it as long prose

Exit condition: When user signals completion ("done", "ready", "that's it", or similar):

•Summarize what you've captured: "I have: 3 successes, 2 mistakes, 1 bottleneck, ratings for mental/productivity/engagement..."
•Ask for confirmation: "Ready to see the structured form, or want to add more?"
•Only proceed to Phase 5 after explicit confirmation

Phase 5: Structured Presentation

•Present the filled-in form with all sections organized:

code

## Time Allocation

### Activities
- Development: 30-40% (Toggl: 35%)
- Meetings: 20-30% (Toggl: 20%)
...

### Projects
- Inspect Action: 40-50%
- Data Pipeline: 20-30%
...

## Goals
✓ Paired with an engineer
✓ Left comments on writeups
✗ Read team writeups
Notes: "Paired with Thomas on the auth bug"

## Reflection
Successes: Got the feature shipped ahead of schedule...
Mistakes: Should have communicated the timeline change earlier...
Bottlenecks: Waiting on PR reviews blocked me for a day...
Action: Set up office hours for quick reviews
Priorities check: Yes, working on the highest impact item
Iteration: Add a daily standup check-in

## Ratings
Mental: 5  Productivity: 6  Prioritization: 5
Time mgmt: 4  Engagement: 5  Overall: 5

## Next Week
Priorities: 1. Ship viewer  2. Onboard Rafael  3. Plan Q1
On track (project): Yes
On track (personal): Unsure

•
Highlight gaps: "Missing: What's one thing you'll change next week?"
•
Show discrepancies: "Your perceived Meeting time (30-40%) vs Toggl (25%)"
•
Ask for edits: "Anything to adjust before we continue?"

Phase 6: Red Team Challenge

Launch the red-teamer agent (using Task tool with subagent_type: "red-teamer").

Context to pass to the agent (include all of this in the prompt):

•Current week data: The complete structured form from Phase 5 (time allocations, reflections, ratings, priorities)
•Historical trends: Summary of last 8 weeks - recurring themes in successes/mistakes/bottlenecks, rating patterns, priority follow-through
•Repeated patterns: Phrases or themes that appear in multiple weeks (e.g., "communicate better" appearing 3 times)

Challenge areas for the agent to probe:

•Compare self-assessment vs objective Toggl data
•Recurring bottlenecks that haven't been addressed
•Priorities that don't align with actual time spent
•Patterns from history: "You've said 'communicate better' for 3 weeks - what's blocking that?"
•Blind spots: activities consuming time without mention

Present challenges to the user. Let them respond and revise if needed.

Phase 7: Save

•
Build the JSON object with all collected data (see schema below)
- •Include red_team with challenges raised and user responses
- •Ensure week.start and week.end are filled (YYYY-MM-DD)
- •Don't abbreviate reflection fields - preserve the user's full prose as they said it
•
Show the user what will be saved before saving
- •Display the reflection fields in full (these are the most important)
- •Ask: "Does this capture everything? Ready to save?"
•
Append to the provided JSONL file
- •Each line is one complete JSON object
- •Append with newline separator
•
Show summary: "Week of Jan 12-18 saved. Review #31."

JSONL Schema

Each line in weekly-reviews.jsonl is a JSON object:

json

{
  "timestamp": "2026-01-19T10:30:00Z",
  "week": {"start": "2026-01-12", "end": "2026-01-18"},
  "activities": {
    "admin": "0-5", "code_review": "10-20", "data": "0",
    "design": "0-5", "development": "40-50", "events": "0",
    "meetings": "10-20", "messages": "0-5", "ops": "0-5",
    "planning": "0-5", "reading": "5-10", "research": "0",
    "training": "0", "writing": "5-10", "other": "0"
  },
  "projects": {
    "3pra": "0", "capabilities": "0", "data_pipeline": "20-30",
    "hiring": "0-5", "human_uplift": "0", "infra": "0-5",
    "inspect_action": "30-40", "managing": "5-10", "middleman": "0",
    "monitorability": "0", "org": "10-20", "tasks_agents": "5-10",
    "vivaria": "0", "other": "0"
  },
  "toggl": {
    "activities": {"development": 35, "meetings": 20, "code_review": 15},
    "projects": {"inspect_action": 40, "data_pipeline": 25}
  },
  "goals": {
    "completed": ["paired_engineer", "read_writeups", "left_comments"],
    "notes": "Paired with Thomas on the auth bug"
  },
  "reflection": {
    "successes": "The foobar feature went really well, users are really happy with it. I think we...",
    "mistakes": "I should have been more patient with Jaime when they asked about timelines. Maybe I could...",
    "bottlenecks": "Waiting on PR reviews blocked me for a day. The auth PR has been open for a week now...",
    "bottleneck_action": "Set up office hours for quick reviews",
    "priorities_check": "Yes, working on the highest impact item",
    "iteration": "Add a daily standup check-in"
  },
  "ratings": {
    "mental": 5, "productivity": 6, "prioritization": 5,
    "time_mgmt": 4, "engagement": 5, "overall": 5
  },
  "next_week": {
    "priorities": ["Ship viewer", "Onboard Rafael", "Plan Q1"],
    "on_track_project": "yes",
    "on_track_personal": "unsure"
  },
  "optional": {
    "prompt": "What surprised me this week?",
    "response": "How much energy I had on Tuesday"
  },
  "red_team": {
    "challenges": "You've mentioned 'communicate better' for 3 weeks now - what's concretely blocking that?",
    "user_responses": "I think I need to add a buffer before responding to pushback..."
  }
}

Field notes:

•reflection.*: Preserve the user's full prose - don't condense "I'm still not prioritizing PMing monitorability enough. I'm not attending stand-ups yet" into "Not prioritizing well". The detail enables future red-teaming.
•red_team: Captures the red team challenges and user's responses for future reference
•
Time semantics - two different representations:
- •activities and projects: User's perceived time allocation (ranges like "0-5", "10-20", "40-50"). These reflect how the user felt they spent their time, adjusted from Toggl data based on narration.
- •toggl.*: Actual recorded time from Toggl (exact integers). Raw data, not adjusted.
- •The gap between these is valuable for red-teaming ("You felt you spent 30% on meetings but Toggl shows 20%")
•goals.completed: Array of checkbox IDs that were checked
•ratings: Integer 1-7 scale
•next_week.on_track_*: "yes", "no", or "unsure"
•optional: For ad-hoc reflection prompts. The prompt field is the question asked, response is the user's answer. Use when exploring specific themes not covered by standard fields.
•week.start and week.end: Always fill these (YYYY-MM-DD format) - don't leave empty

Fallback Handling

Toggl Failure

If toggl_get_time_entries is unavailable or fails:

•Inform the user: "Toggl integration unavailable - we'll do manual time entry"
•Skip Phase 2 entirely
•In Phase 3, present empty time allocation tables
•Ask user to estimate percentages directly: "How did you spend your time this week? Estimate percentages for your main activities."

Chart Failure

If plotext fails or produces garbled output:

•Fall back to text-based visualization:

code

Development:  ████████████████████ 40%
Meetings:     ██████████ 20%
Code Review:  ███████ 15%
Writing:      █████ 10%
Other:        ███████ 15%

•Use Unicode block characters (█) scaled to percentage
•Still show trend indicators: "↗ up from 30%" or "↘ down from 50%"

Edge Cases

•Missing JSONL file: Start fresh, no historical trends to show
•Malformed JSON lines: Skip and warn, continue with valid data
•No Toggl data: Show empty tables, proceed with manual entry (see Fallback Handling)
•User skips sections: Fill with empty values, note gaps
•Not Sunday: Warn but allow running anyway ("Running mid-week - date range may be unexpected")

Goal Tracking Checkboxes

Standard goals to track (user can mention any of these):

•paired_engineer: Paired with an engineer this week
•read_writeups: Read team members' writeups
•left_comments: Left comments on others' work
•gave_feedback: Gave constructive feedback
•asked_help: Asked for help when stuck
•documented: Documented decisions or learnings
•mentored: Mentored or helped onboard someone
•shipped: Shipped something to users

Notes

•Uses toggl_get_time_entries MCP tool for Toggl integration
•Red team uses Task tool with subagent_type: "red-teamer"
•Timezone: Default to US Pacific. Query Toggl with expanded UTC range, then filter to local timezone.
•Charts: Use uvx --with plotext python for terminal bar charts (no install needed)
•JSONL format allows easy appending and parsing