todoist-manage

Todoist Management Skill

Complete personal productivity manager integrating Todoist with Claude Code for natural-language task management, strategic planning, and productivity coaching.

Prerequisites

1. Check Environment Variable

echo $TODOIST_TOKEN

If not set:

❌ TODOIST_TOKEN not set. You need a Todoist API token to use this skill.

How to get your token:
1. Go to https://todoist.com/app/settings/integrations/developer
2. Scroll to "API token" section
3. Copy your token
4. Run: export TODOIST_TOKEN="your_token_here"
5. Add to your shell profile (~/.zshrc or ~/.bashrc) to persist

2. Check SDK Installation

python -c "import todoist_api_python; print('✓ SDK installed')"

If not installed:

pip install 'todoist-api-python>=3.1.0,<4.0.0'
# or
uv pip install 'todoist-api-python>=3.1.0,<4.0.0'

3. Verify Script Access

Script location: plugins/todoist-manager/skills/todoist-manage/scripts/todoist_client.py

Test:

python3 plugins/todoist-manager/skills/todoist-manage/scripts/todoist_client.py overview

Expected: JSON output with overview data

Script Usage

All operations use the todoist_client.py script with JSON output.

General Pattern

python3 <script_path> <resource> <action> [--options]

Resources: tasks, projects, sections, labels, comments, overview, daily-summary

Output: Always JSON with this structure:

{
  "success": true,
  "data": { ... },
  "count": 5
}

Or for errors:

{
  "error": true,
  "message": "Error description",
  "context": { ... }
}

Section A: Task Management

List All Tasks

python3 todoist_client.py tasks list

Options:

• --project-id 12345 - Filter by project
• --section-id 67890 - Filter by section
• --label urgent - Filter by label name
• --ids 111 222 333 - Specific task IDs
• --include-child-projects - Include tasks from all child/descendant projects (requires --project-id)

Filter Tasks with Queries

python3 todoist_client.py tasks filter --query "today & p1"

Common queries:

• "today" - Tasks due today
• "overdue" - All overdue tasks
• "7 days" - Tasks due in next 7 days
• "today & p1" - Urgent tasks due today
• "#Work" - Tasks in Work project
• "@urgent" - Tasks with urgent label
• "today & !@routine" - Today's tasks excluding routine
• "(overdue | today) & (p1 | p2)" - High-priority tasks needing attention

See: references/filter-query-syntax.md for complete query language

Get Single Task

python3 todoist_client.py tasks get --task-id 12345

Add Task

Full form:

python3 todoist_client.py tasks add \
  --content "Task title" \
  --description "Additional details" \
  --project-id 12345 \
  --section-id 67890 \
  --labels urgent important \
  --priority 4 \
  --due-string "tomorrow at 3pm" \
  --duration 30 \
  --duration-unit minute

Minimal:

python3 todoist_client.py tasks add --content "Buy milk"

Priority values:

• 1 = Normal (P4 in UI)
• 2 = High (P3 in UI)
• 3 = Higher (P2 in UI)
• 4 = Urgent (P1 in UI)

Due date options (use ONE):

• --due-string "tomorrow at 3pm" - Natural language (recommended)
• --due-date "2025-02-24" - Date only (YYYY-MM-DD)
• --due-datetime "2025-02-24T15:00:00Z" - Full datetime (RFC3339)

Duration for time-blocking:

• --duration 30 --duration-unit minute - 30-minute task
• --duration 2 --duration-unit day - 2-day task

Quick Add Task

python3 todoist_client.py tasks quick-add --text "Buy milk tomorrow @shopping p1"

Quick-add syntax:

• tomorrow - Due date
• @shopping - Label
• p1 - Priority (urgent)
• #Work - Project
• !3 - Priority (alternative syntax)
• every monday - Recurring
• ^today - Due date with time
• //Section Name - Section

Examples:

• "Call John tomorrow at 2pm @phone p2 #Work"
• "Team meeting every Friday at 10am #Work"
• "Review PR !1 ^today"

Update Task

python3 todoist_client.py tasks update \
  --task-id 12345 \
  --content "Updated title" \
  --priority 4 \
  --due-string "friday"

Updatable fields: content, description, labels, priority, due (date/string/datetime), duration

Complete Task

python3 todoist_client.py tasks complete --task-id 12345

Uncomplete Task (Reopen)

python3 todoist_client.py tasks uncomplete --task-id 12345

Delete Task

python3 todoist_client.py tasks delete --task-id 12345

⚠️ Confirm before deleting unless clearly temporary or user explicitly requested deletion.

Bulk Operations

For multiple tasks, use parallel bash calls or loop:

# Complete multiple tasks
for task_id in 111 222 333; do
  python3 todoist_client.py tasks complete --task-id $task_id
done

Present summary:

✅ Completed 3 tasks:
- Task A
- Task B
- Task C

Section B: Project & Organization

Projects

List all:

python3 todoist_client.py projects list

Get one:

python3 todoist_client.py projects get --project-id 12345

Add project:

python3 todoist_client.py projects add \
  --name "New Project" \
  --color "blue" \
  --is-favorite \
  --view-style "board"

Colors: berry_red, red, orange, yellow, olive_green, lime_green, green, mint_green, teal, sky_blue, light_blue, blue, grape, violet, lavender, magenta, salmon, charcoal, grey, taupe

Update:

python3 todoist_client.py projects update --project-id 12345 --name "Renamed Project"

Delete:

python3 todoist_client.py projects delete --project-id 12345

⚠️ Confirm before deleting projects - this deletes all contained tasks.

Sections

List (all or by project):

python3 todoist_client.py sections list
python3 todoist_client.py sections list --project-id 12345

Add section:

python3 todoist_client.py sections add --name "In Progress" --project-id 12345 --order 1

Update:

python3 todoist_client.py sections update --section-id 67890 --name "Completed"

Delete:

python3 todoist_client.py sections delete --section-id 67890

Labels

List all:

python3 todoist_client.py labels list

Add label:

python3 todoist_client.py labels add \
  --name "urgent" \
  --color "red" \
  --is-favorite

Update:

python3 todoist_client.py labels update --label-id 54321 --name "super-urgent" --color "berry_red"

Delete:

python3 todoist_client.py labels delete --label-id 54321

Comments

List (for task or project):

python3 todoist_client.py comments list --task-id 12345
python3 todoist_client.py comments list --project-id 67890

Add comment:

python3 todoist_client.py comments add --content "Progress update: 50% complete" --task-id 12345

Update:

python3 todoist_client.py comments update --comment-id 99999 --content "Updated progress: 75% complete"

Delete:

python3 todoist_client.py comments delete --comment-id 99999

Use comments for:

• Progress updates on long-running tasks
• Blockers or notes
• Reminders from user to themselves
• Context for future reference

Section C: Daily Planning

Morning Review Workflow

Goal: Start the day with clarity and focus.

1. Fetch overview:

python3 todoist_client.py overview

Output structure:

{
  "success": true,
  "data": {
    "overdue": {
      "count": 3,
      "by_project": {
        "Work": [...],
        "Personal": [...]
      }
    },
    "today": {
      "count": 12,
      "by_project": {...}
    },
    "upcoming": {
      "count": 25,
      "by_project": {...}
    },
    "total_pending": 40
  }
}

2. Present with priority grouping:

📊 Good morning! Here's your overview:

🔴 Overdue (3)
  Work:
    □ [P1] Fix production bug (urgent!)
    □ [P2] Client proposal draft
  Personal:
    □ [P3] Schedule dentist appointment

📅 Today (12 tasks)
  Work (8):
    □ [P1] Team standup (9am)
    □ [P1] Deploy feature X
    □ [P2] Code review for PR #123
    □ [P2] Update documentation
    □ [P3] Clean up backlog
    ...
  Personal (4):
    □ [P2] Gym workout
    □ [P3] Meal prep
    ...

📆 Upcoming (25 tasks in next 7 days)
  Top priorities:
    - [P1] Quarterly planning (Friday)
    - [P1] Client demo (Thursday)
    - [P2] 1-on-1 with manager (Wednesday)

3. Coach for focus selection:

Ask user:

• "Which 3-5 tasks would make today a success?"
• "What's your energy level today? (High/Medium/Low)"
• "Any blockers or dependencies?"
• "Expected meetings or interruptions?"

4. Generate time-blocked schedule:

Based on user selections and task durations:

⏰ Your Focus Plan for Today

🌅 Morning (9am-12pm) - Deep Work
  □ 9:00-9:15   Team standup (15 min)
  □ 9:30-11:00  Deploy feature X (90 min) [P1] 🔴
  □ 11:00-12:00 Code review PR #123 (60 min) [P2]

☀️ Afternoon (1pm-5pm) - Collaboration
  □ 1:00-2:00   Update documentation (60 min) [P2]
  □ 2:00-3:00   Client proposal draft (60 min) [P2] ⚠️ overdue
  □ 3:00-4:00   Fix production bug (60 min) [P1] 🔴 ⚠️ overdue

🌙 Evening (5pm-6pm) - Personal
  □ 5:00-6:00   Gym workout (60 min) [P2]

⏱️  Total: 6.25 hours (buffer: 30 min) ✅

💡 Tips:
- You have 3 overdue tasks - prioritizing those first
- Schedule looks achievable with buffer
- Consider gym as optional if running behind

5. Create focus filter:

# Save user's focus selections as custom filter
python3 todoist_client.py tasks filter --query "(id:111 | id:222 | id:333 | id:444 | id:555)"

Or create a label:

python3 todoist_client.py labels add --name "today-focus"
# Then add label to selected tasks
python3 todoist_client.py tasks update --task-id 111 --labels today-focus work

6. Proactive suggestions:

• Too many P1 tasks (>5)? "You have 8 urgent tasks today. That's overwhelming. Can we reschedule some or renegotiate deadlines?"
• Long tasks without subtasks? "This 3-hour task looks big. Want to break it into smaller steps?"
• Overdue tasks? "You have 3 overdue tasks. Let's triage - which are still relevant?"
• Over capacity? "You have 8 hours of tasks scheduled with only 6 hours available. Let's defer some to later this week."

See: references/productivity-workflows.md for complete morning review template

Evening Review Workflow

Goal: Reflect on progress and prep for tomorrow.

1. Fetch daily summary:

python3 todoist_client.py daily-summary

Output:

{
  "success": true,
  "data": {
    "completed_today": {
      "count": 8,
      "tasks": [...]
    },
    "remaining_today": {
      "count": 4,
      "tasks": [...]
    },
    "completion_rate": "66.7%"
  }
}

2. Celebrate wins:

🎉 Great work today!

✅ Completed: 8/12 tasks (67%)

What you accomplished:
  ✓ Deploy feature X (P1) - Nice work!
  ✓ Team standup
  ✓ Code review PR #123 (P2)
  ✓ Update documentation
  ✓ Gym workout
  ...

🌟 You cleared 2 overdue tasks and completed all P1s!

3. Review incomplete tasks:

⏳ Remaining tasks (4):

□ Client proposal draft (P2) - Due today
  → Reschedule to tomorrow?

□ Fix production bug (P1) - Due today
  → What blocked this?

□ Clean up backlog (P3)
  → Still relevant or defer?

□ Meal prep (P3)
  → Move to weekend?

For each incomplete, ask:

• "Still relevant?"
• "What blocked completion?"
• "New due date?"
• "Should we break this down?"

Actions:

• Reschedule: python3 todoist_client.py tasks update --task-id 123 --due-string "tomorrow"
• Defer: --due-string "next week" or remove due date
• Delete: python3 todoist_client.py tasks delete --task-id 123
• Break down: Create subtasks

4. Prep for tomorrow:

python3 todoist_client.py tasks filter --query "tomorrow"

Present tomorrow's overview:

📅 Tomorrow Preview (8 tasks)

Top priorities:
  □ [P1] Client proposal draft (rescheduled from today)
  □ [P1] Quarterly planning prep
  □ [P2] 1-on-1 with manager

Looks manageable! Get good rest. 😴

5. Flag patterns:

• Task deferred 3+ times? "I notice 'Client proposal' keeps getting rescheduled. Is this still a priority?"
• Consistently incomplete days? "You've had 4 days in a row under 50% completion. Feeling overwhelmed?"
• Same tasks always late? "Your gym workouts keep slipping. Want to move them to morning?"

See: references/productivity-workflows.md for complete evening review template

Weekly Review Workflow

Goal: Big-picture perspective and course correction.

1. Generate weekly metrics:

# Approximate with filters
python3 todoist_client.py tasks filter --query "completed this week"
python3 todoist_client.py tasks filter --query "next 7 days"
python3 todoist_client.py tasks filter --query "overdue"

2. Present weekly dashboard:

📊 Week of Feb 17-23, 2025

Completed: 45 tasks ✅
Overdue: 3 tasks ⚠️
Completion rate: 82% 🎯

By Project:
  Work: 28 tasks (62%)
  Personal: 15 tasks (33%)
  Side Project: 2 tasks (5%)

By Priority:
  P1: 15 tasks (33%)
  P2: 20 tasks (44%)
  P3+: 10 tasks (22%)

Project Health:
  🟢 Work - On track (8 tasks/week, low overdue)
  🟡 Side Project - At risk (no activity in 5 days)
  🔴 Personal - Stalled (10 tasks overdue, no progress)

3. Review project health:

For each project:

• ✅ Completed tasks this week
• ⚠️ Overdue count
• 🕐 Days since last activity
• 📋 Tasks without due dates

Status indicators:

• 🟢 On track: Progress, low overdue
• 🟡 At risk: High overdue OR no activity 5-7 days
• 🔴 Stalled: No activity 7+ days

4. Coaching questions:

• "What went well this week?"
• "What blocked progress on [Stalled Project]?"
• "Side Project has had no activity - pause or delete?"
• "What's the top priority for next week?"
• "Do you need a lighter week or can you push harder?"

5. Identify patterns:

• Priority inflation: "75% of tasks are P1/P2 - that's not sustainable"
• Project explosion: "Work project has 50 tasks - needs organization"
• Label neglect: "Only 20% of tasks have context labels"
• Duration missing: "Most tasks have no time estimates - hard to plan"

6. Weekly maintenance:

• Archive completed projects
• Delete cancelled tasks
• Update stale due dates
• Review "no date" inbox
• Clean up unused labels
• Organize sections

See: references/productivity-workflows.md for complete weekly review template

Section D: Productivity Coach

Task Breakdown

When to trigger:

• Task duration >60 minutes
• User says "I don't know where to start"
• Task has been deferred 2+ times
• Vague task content ("Work on project X")

Process:

1. Clarify outcome:

• "What does 'done' look like?"
• "What's the deliverable?"
• "Who needs what from this?"

2. Identify steps:

• Each step starts with action verb
• Each step <30 minutes
• Each step is concrete and actionable

3. Create subtasks:

# Add parent task if needed
python3 todoist_client.py tasks add \
  --content "Write quarterly report" \
  --project-id 12345 \
  --due-string "Friday"

# Get task ID from response, then add subtasks
python3 todoist_client.py tasks add \
  --content "Research last quarter's data" \
  --parent-id [parent_task_id] \
  --duration 30 \
  --duration-unit minute \
  --labels computer \
  --priority 2

python3 todoist_client.py tasks add \
  --content "Draft executive summary" \
  --parent-id [parent_task_id] \
  --duration 45 \
  --duration-unit minute \
  --labels computer focus \
  --priority 2

# Continue for all subtasks

4. Add context:

• Duration estimates (--duration, --duration-unit)
• Labels for context (@computer, @phone, @office)
• Labels for energy (@focus, @quick, @routine)
• Due dates for sequencing

Example breakdown:

"Write quarterly report" →

1. Research last quarter's data (30 min) @computer
2. Draft executive summary (45 min) @focus @computer
3. Create performance charts (30 min) @computer
4. Write detailed sections (90 min) @focus @computer
5. Review and edit (20 min) @computer
6. Get feedback from manager (async) @waiting
7. Incorporate feedback (30 min) @computer
8. Final proofread (15 min) @computer
9. Submit report (5 min) @computer

Total: ~4.5 hours of work time + async feedback

See: references/productivity-workflows.md for complete breakdown framework

Time-Boxing

When to use:

• User asks "will this fit today?"
• Daily plan exceeds available time
• User wants realistic schedule

Process:

1. Calculate available time:

Ask:

• "What time do you need to stop today?"
• "How many meetings do you have?"
• "Any committed blocks?"

Calculate:

Available = End - Start - Meetings - Breaks - Buffer (10%)

Example:
9am-6pm = 9 hours = 540 minutes
- Meetings: 2 hours = 120 minutes
- Lunch + breaks: 1 hour = 60 minutes
- Buffer (10%): 36 minutes
= Available: 324 minutes (5.4 hours)

2. Fetch tasks with durations:

python3 todoist_client.py tasks filter --query "today & !no duration"

3. Prioritize and fit:

Algorithm:

1. Sort by priority (P1 > P2 > P3 > P4)
2. Sum P1 task durations
3. If under capacity, add P2 tasks
4. Continue until 80-90% capacity

4. Present schedule:

📅 Today's Schedule (324 min available)

🔴 Must Do (180 min)
  □ 9:00-10:30  Deploy feature X (90 min)
  □ 11:00-12:00 Code review (60 min)
  □ 2:00-2:30   Client call (30 min)

🟡 Should Do (90 min)
  □ 2:30-3:30   Documentation (60 min)
  □ 3:30-4:00   Email responses (30 min)

Buffer: 54 minutes ✅

⚠️ Deferred to tomorrow:
  □ Backlog cleanup (60 min) - P3
  □ Learning time (45 min) - P3

Capacity warnings:

• ⚠️ Over capacity: Total >100% available
• ⚠️ No buffer: Total = 95-100% available
• ✅ Good balance: Total = 80-90% available
• 😌 Light day: Total <70% available

5. Match energy levels:

Morning (9am-12pm): Deep work

• @focus tasks
• Long tasks (>60 min)
• Creative/strategic work

Afternoon (1pm-5pm): Collaboration

• @collaborate tasks
• Meetings
• Communication

Evening (5pm-6pm): Admin

• @quick tasks
• <30 minute tasks
• Email, planning

# Morning focus tasks
python3 todoist_client.py tasks filter --query "today & @focus & duration > 60"

# Evening quick tasks
python3 todoist_client.py tasks filter --query "today & @quick & duration < 30"

See: references/productivity-workflows.md for complete time-boxing framework

Pattern Recognition

Automatically detect and flag:

🟢 Green Flags (Healthy):

• Completion rate >70%
• Balanced priorities
• Regular progress on all projects
• Minimal deferrals
• Consistent reviews

🟡 Yellow Flags (Warning):

• Completion rate 50-70%
• Priority inflation (too many P1s)
• Frequent rescheduling
• Growing "no date" backlog
• Inconsistent project activity

🔴 Red Flags (Action Needed):

• Completion rate <50%
• Chronic overdue (>10 tasks)
• Project stalled 7+ days
• Task deferred 5+ times
• Daily load >8 hours
• No reviews in 7+ days

Coaching interventions:

For priority inflation:

"I notice 15 tasks marked P1. That's overwhelming - not everything can be urgent.
Let's identify the true must-dos for today and downgrade or defer the rest."

For chronic deferrals:

"This task has been rescheduled 5 times. Let's figure out why:
- Still relevant? Or delete?
- Too big? Break it down?
- Blocked? Create unblocking task?
- Lost motivation? Defer indefinitely?"

For overload:

"You have 12 hours of tasks for 6 hours available time.
Let's prioritize the critical 3-4 tasks and defer the rest."

See: references/productivity-workflows.md for complete pattern recognition guide

Proactive Suggestions

Always offer when relevant:

After showing tasks:

• "Want me to estimate durations for time-blocking?"
• "Should we add context labels (@office, @computer, @home)?"
• "Any tasks feel too big to tackle?"
• "Which is most critical for today?"

During planning:

• "This task has been on your list 3 weeks - still want it?"
• "You have 5 projects with P1 tasks - too many priorities?"
• "All tasks in one project - want to organize by area?"

After completion:

• "Nice work! 3/5 P1 tasks done."
• "Great momentum - overdue list is clear!"
• "What would you like to tackle next?"
• "You have 90 minutes left - perfect for [Task Y]."

Encourage habits:

• "Would daily reviews at 8am help?"
• "Want a recurring weekly review every Sunday?"
• "Should we create a morning routine checklist?"

Error Handling

Missing Token

{
  "error": true,
  "message": "TODOIST_TOKEN environment variable not set"
}

Response:

❌ Todoist token not configured.

To fix:
1. Get your token: https://todoist.com/app/settings/integrations/developer
2. Set it: export TODOIST_TOKEN="your_token_here"
3. Persist: Add to ~/.zshrc or ~/.bashrc
4. Reload: source ~/.zshrc

Then try again!

Rate Limiting

{
  "error": true,
  "message": "Failed to list tasks: 429 Too Many Requests"
}

Response:

⚠️ Hit Todoist rate limit (too many requests).

The API has automatic retry, but you might need to wait a minute.
Try again in 60 seconds.

Tip: Use filter queries instead of listing all tasks to reduce API calls.

Invalid Task ID

{
  "error": true,
  "message": "Failed to get task: Task not found",
  "task_id": "12345"
}

Response:

❌ Task 12345 not found. It may have been deleted or the ID is incorrect.

Want to search for the task instead? I can filter by content.

Network Errors

{
  "error": true,
  "message": "Failed to list tasks: Connection timeout"
}

Response:

❌ Can't reach Todoist API. Check your internet connection.

If you're online, Todoist might be experiencing issues.
Status: https://status.todoist.com

Output Formatting

Task Lists

Format based on context:

As checklist:

□ Task A (P1) - Due today
□ Task B (P2) - Due tomorrow
□ Task C (P3) - No due date

As table:

| Task                | Priority | Due Date  | Project |
|---------------------|----------|-----------|---------|
| Deploy feature X    | P1 🔴    | Today     | Work    |
| Code review         | P2 🟡    | Tomorrow  | Work    |
| Gym workout         | P2 🟡    | Today     | Personal|

Grouped by project:

Work (8 tasks):
  🔴 P1: Deploy feature X (due today)
  🟡 P2: Code review (due tomorrow)
  ⚪ P3: Documentation update (no date)

Personal (4 tasks):
  🟡 P2: Gym workout (due today)
  ⚪ P3: Meal prep (due Sunday)

Priority Indicators

• 🔴 P1 (Urgent) - priority = 4
• 🟡 P2 (Higher) - priority = 3
• 🟠 P3 (High) - priority = 2
• ⚪ P4 (Normal) - priority = 1

Status Indicators

• ✅ Completed
• □ Incomplete
• ⚠️ Overdue
• ⏰ Due today
• 📅 Upcoming
• 🔄 Recurring
• ⏳ In progress (has comments/activity)

Contextual Formatting

For morning review: Group by priority, highlight overdue For time-blocking: Format as timeline with durations For weekly review: Show metrics and trends For search results: Show relevant fields (due date, project, priority)

Always:

• Use emojis for visual clarity
• Highlight urgent/overdue items
• Group logically (by project, by priority, by date)
• Show counts for awareness
• Provide actionable next steps

Safety Guidelines

Always Confirm

Destructive actions:

• Deleting tasks (unless temporary)
• Deleting projects
• Deleting labels
• Bulk deletions (>5 items)

Prompt: "This will delete [X]. This can't be undone. Confirm?"

Usually Confirm

Significant changes:

• Moving tasks between projects
• Changing due dates by >7 days
• Completing tasks on behalf of user
• Bulk updates (>10 items)
• Archiving projects

Prompt: "This will [action] affecting [N] items. Proceed?"

No Confirmation Needed

Read operations:

• Listing tasks
• Filtering
• Viewing overviews
• Searching

Minor updates:

• Adding tasks
• Updating content/description
• Adding labels
• Adjusting due date by 1-2 days
• Creating subtasks
• Adding comments

Trust Your Judgment

When unclear, err on side of confirmation for user peace of mind.

Exceptions:

• User explicitly said "delete all overdue tasks" → Confirm even though explicit
• User is in flow state completing many tasks → Skip confirmations for completions
• Emergency situation ("delete everything!") → Still confirm once

Reference Documents

• references/todoist-api-reference.md - Complete SDK method reference, parameters, return types
• references/filter-query-syntax.md - Full Todoist filter query language with examples
• references/productivity-workflows.md - Strategic planning templates, coaching frameworks

When to consult:

• Need to verify exact API parameter names → todoist-api-reference.md
• Building complex filter query → filter-query-syntax.md
• Planning daily/weekly reviews → productivity-workflows.md
• Coaching on productivity patterns → productivity-workflows.md

Session Examples

Example 1: Morning Planning

User: "Good morning, let's plan my day"

Claude:

1. Runs python3 todoist_client.py overview

2. Parses JSON output

3. Presents with priority grouping:

   📊 Good morning! Here's your overview:
   
   ⚠️ 3 overdue tasks (2 urgent)
   📅 12 tasks due today
   📆 25 upcoming in next 7 days
   
   Top priorities for today:
   🔴 [P1] Fix production bug (overdue)
   🔴 [P1] Deploy feature X
   🔴 [P1] Client call at 2pm
   
   This looks achievable. Which 3-5 tasks would make today a success?
   

4. User selects focus tasks

5. Generates time-blocked schedule

6. Creates focus filter or label

7. Estimates total time vs available

8. Offers to defer if over capacity

Example 2: Task Breakdown

User: "I need to work on the client proposal but don't know where to start"

Claude:

1. Asks clarifying questions:

   Let me help break this down.
   
   - What's the proposal for? (Service offering? Pricing? Pitch?)
   - Who's the audience? (Technical? Executive?)
   - What's the deadline?
   - How long do you think this will take total?
   

2. User answers

3. Suggests breakdown:

   Here's a breakdown:
   
   1. Research client's needs (30 min) @computer
   2. Draft outline (20 min) @computer
   3. Write executive summary (45 min) @focus @computer
   4. Detail service offering (60 min) @focus @computer
   5. Create pricing table (30 min) @computer
   6. Design cover page (20 min) @computer
   7. Review and polish (20 min) @computer
   8. Get internal feedback (async) @waiting
   9. Incorporate feedback (30 min) @computer
   10. Send to client (5 min) @computer
   
   Total: ~4 hours of work + async feedback
   
   Should I create these as subtasks in Todoist?
   

4. User confirms

5. Creates parent task + all subtasks with appropriate metadata

6. Suggests starting with research (first subtask)

Example 3: Evening Review

User: "Let's review my day"

Claude:

1. Runs python3 todoist_client.py daily-summary

2. Parses output

3. Celebrates wins:

   🎉 Solid day!
   
   ✅ Completed: 8/12 tasks (67%)
   
   What you accomplished:
   ✓ Deploy feature X (P1) - Big win! 🚀
   ✓ Client call
   ✓ Code review
   ✓ Documentation update
   ✓ Gym workout
   ✓ Email responses
   ✓ Team standup
   ✓ Planning session
   
   You cleared all your P1 tasks and made progress on 2 projects!
   

4. Reviews incomplete tasks:

   ⏳ Remaining (4 tasks):
   
   □ Fix production bug (P1) - Due today
     What blocked this? Still urgent?
   
   □ Client proposal (P2) - Due today
     Want to reschedule to tomorrow?
   
   □ Backlog cleanup (P3)
     Defer to this weekend?
   
   □ Meal prep (P3)
     Move to tomorrow or Sunday?
   

5. For each, gets user input and takes action (reschedule, defer, delete)

6. Prepares tomorrow's overview

7. Flags any patterns (if applicable)

Quick Reference Commands

Essential operations:

# Morning overview
python3 todoist_client.py overview

# Today's tasks
python3 todoist_client.py tasks filter --query "today"

# Add task
python3 todoist_client.py tasks add --content "Task name" --due-string "tomorrow"

# Quick add (natural language)
python3 todoist_client.py tasks quick-add --text "Buy milk tomorrow @shopping p1"

# Complete task
python3 todoist_client.py tasks complete --task-id 12345

# Filter (custom query)
python3 todoist_client.py tasks filter --query "today & p1"

# Daily summary
python3 todoist_client.py daily-summary

# List projects
python3 todoist_client.py projects list

# List labels
python3 todoist_client.py labels list

Script path: plugins/todoist-manager/skills/todoist-manage/scripts/todoist_client.py

Integration with Agent

For complex multi-turn workflows (weekly planning, major reorganization, strategic coaching), spawn the todoist-manager agent:

# Example trigger
"I need help completely reorganizing my task list and projects"

The agent has full access to this skill and can conduct extended planning sessions with strategic thinking.

When to use agent:

• Multi-day/week planning sessions
• Complete task list reorganization
• Strategic project planning
• Extended productivity coaching
• Complex goal setting

When to use skill directly:

• Quick task operations
• Daily/evening reviews
• Single task breakdowns
• Viewing overviews
• Routine task management