task-review

Task Readiness Review Skill

Structured review of 1-3 Todoist tasks for clarity, actionability, and execution readiness. Scores each task across 5 dimensions and provides specific improvement recommendations.

When to Use This Skill

Use for focused reviews of 1-3 tasks when the user wants:

• To understand if a task is ready to execute
• Specific feedback on task quality
• Recommendations to improve task clarity

For batch reviews of many tasks, use the todoist-manager agent instead.

Prerequisites

• TODOIST_TOKEN set in environment
• todoist-api-python SDK installed (auto-installs if missing)
• Script: plugins/todoist-manager/skills/todoist-manage/scripts/todoist_client.py

Review Process

Step 1: Resolve Input to Task(s)

The input can be a task ID, task content, or a project name. Resolve in this order:

1a. Task ID provided — fetch directly:

python3 <script_path> tasks get --task-id <ID>

1b. Looks like a project name — check projects list first, then use ## to fetch all tasks including sub-projects:

# List projects to find a name match
python3 <script_path> projects list

# If matched, fetch all tasks in project + sub-projects
python3 <script_path> tasks filter --query "##<Project Name>"

1c. Task content provided — fall back to text search:

python3 <script_path> tasks filter --query "search: <task content>"

Resolution priority: task ID > project name match > text search. Always prefer project-based lookup when the input matches a project name, because text search only finds tasks where the search string appears in the task title.

Step 2: Score the Task

Score each dimension using the readiness rubric (references/readiness-rubric.md):

Dimension	Max Score	Key Question
Clarity	2	Is the task content unambiguous?
Actionability	2	Does it start with a clear action verb?
Scope	2	Can it be done in one session (<2 hours)?
Context	2	Does it have the right metadata?
Outcome	2	Is "done" clearly defined?

Total: 10 points

Step 3: Present Scorecard

Format:

📋 Task Review: "[Task Content]"
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Clarity     ██████░░░░ 3/5  ⚠️
Actionability ██████████ 5/5 ✅
Scope       ████░░░░░░ 2/5  ❌
Context     ████████░░ 4/5  ✅
Outcome     ██████░░░░ 3/5  ⚠️

Overall: 17/25 — Needs Improvement

Issues Found:
❌ Scope too large — estimate suggests 3-4 hours
⚠️ Missing due date for time-sensitive work
⚠️ Outcome not clearly defined

Recommendations:
1. Break into 2-3 subtasks (30-60 min each)
2. Add due date: --due-string "friday"
3. Add success criterion to description

Want me to apply these improvements? [Y/N]

Step 4: Apply Improvements (with confirmation)

Based on review, offer to:

Update task content:

python3 <script_path> tasks update \
  --task-id <ID> \
  --content "Verb + specific action + outcome" \
  --description "Done when: [criterion]" \
  --due-string "friday" \
  --labels focus computer

Break into subtasks:

# Create subtasks from parent
python3 <script_path> tasks add \
  --content "Step 1: Research X" \
  --parent-id <ID> \
  --duration 30 \
  --duration-unit minute

Add context labels:

python3 <script_path> tasks update \
  --task-id <ID> \
  --labels focus computer

Scoring Reference

See references/readiness-rubric.md for the complete 5-dimension scoring rubric with detailed criteria.

See references/task-templates.md for 7 task type templates (research, writing, coding, meeting, admin, communication, planning).

Review Output Formats

Quick Review (single task)

✅ Task Ready: "Write unit tests for auth module"
Score: 22/25 — Ready to Execute

Minor suggestions:
- Consider adding @focus label for deep work
- Due date is 5 days out — still realistic?

Detailed Review (task needing work)

📋 Task Review: "Work on the project"
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

❌ NOT READY — Score: 8/25

Critical Issues:
1. Clarity (1/5): "Work on the project" is too vague
   → Which project? What aspect? What outcome?

2. Actionability (1/5): No action verb
   → Start with: Research / Write / Review / Build / Fix / Test

3. Scope (2/5): "Project" implies undefined scope
   → What's the smallest meaningful next step?

4. Context (2/5): No project, labels, or due date
   → Add context to make it filterable

5. Outcome (2/5): No success criterion
   → When is this "done"?

Revised task suggestion:
"Draft project requirements document for Q2 feature"
+ Description: "Done when: 1-page requirements doc reviewed by stakeholder"
+ Due: next Tuesday
+ Labels: @focus @computer
+ Duration: 90 min

Apply this revision? [Y/N]

Batch Review (2-3 tasks)

📋 Reviewing 3 Tasks for Readiness
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

✅ Task 1: "Write unit tests for auth module" — 22/25 Ready
⚠️  Task 2: "Update documentation" — 14/25 Needs Work
❌ Task 3: "Handle the client thing" — 6/25 Not Ready

Detailed reviews follow...

Improvement Patterns

Vague → Specific

Before	After
"Work on project"	"Write API endpoint for user login"
"Do the report"	"Draft Q2 metrics report (2 pages)"
"Fix the bug"	"Fix null pointer in user.profile.get()"
"Call someone"	"Call Sarah re: contract renewal"

Missing Action Verb

Strong action verbs: Research, Write, Review, Build, Fix, Test, Deploy, Update, Draft, Design, Schedule, Send, Call, Analyze, Create, Document

Scope Reduction

If task > 2 hours:

1. Identify the smallest meaningful next step
2. Make that the task
3. Add remaining steps as subtasks

Context Labels

Energy context:

• @focus — requires deep concentration
• @quick — under 15 minutes
• @routine — habitual, low-energy

Location context:

• @computer — requires desktop/laptop
• @phone — can do from phone
• @office — requires physical presence
• @home — home environment needed

Status context:

• @waiting — blocked on someone else
• @blocked — has an explicit blocker

Safety

• Always show proposed changes before applying
• Never delete task content without confirmation
• Preserve existing labels/metadata when adding new ones
• If breaking into subtasks, keep parent task as container