Scanneur de Clarté des Tâches

VérifiéSûr

Analyse vos notes quotidiennes avec structure Kanban personnel. Clarifie les tâches, gère le flux Today's 3 ↔ Ready, signale les éléments obsolètes et aide à organiser votre tableau.

Spar Skills Guide Bot
ProductiviteIntermédiaire
03/03/2026
Claude Code
#task-management#personal-kanban#obsidian#daily-notes#prioritization

Recommandé pour

Synthèse automatique

Ce skill analyse une note quotidienne structurée en Kanban personnel pour clarifier les tâches vagues, gérer les colonnes Today's 3 et Ready, signaler les éléments stagnants et mettre à jour le fichier après validation.

Points forts

  • Applique une limite WIP explicite pour Today's 3 et propose des arbitrages quand la colonne déborde.
  • Traite les tâches une par une avec des options concrètes : reformuler, accepter une suggestion, reporter, créer un fichier projet ou déplacer vers Ready.
  • Gère les tâches stagnantes avec des choix clairs : faire, déléguer, abandonner, reformuler ou parquer en someday.
  • Suit un flux en deux passes : diagnostic du tableau puis clarification avant toute modification.

Limites

  • Suppose une structure de note et un chemin Obsidian précis, avec un exemple de chemin personnel codé en dur.
  • S'appuie sur des conventions de dates et de dossiers qui peuvent ne pas correspondre à une autre organisation.
  • Les modifications ne sont pas automatiques : elles nécessitent l'approbation de l'utilisateur et des outils d'édition fonctionnels.
Quand l'utiliser

À utiliser lorsqu'il faut faire le point sur une note quotidienne Kanban, clarifier des tâches floues ou rééquilibrer Today's 3 et Ready.

Quand l'éviter

À éviter si les tâches ne sont pas dans une note quotidienne structurée en Kanban personnel ou si aucun fichier Obsidian compatible n'est accessible.

Analyse de sécurité

Sûr
Score qualité78/100

The skill only uses Read/Glob/Grep/Edit/Write/AskUserQuestion against local Obsidian markdown notes and requires explicit user approval before any edit; it contains no shell execution, network calls, credential handling, or destructive file operations. The only notable quirk is a hardcoded personal path (/Users/eddale/Documents/COPYobsidian/...), which is a portability/privacy nit rather than an execution risk. The optional "PASS 3: Execute" section is gated behind user approval and does not define any dangerous behavior.

Points d'attention
  • •Writes (Edit/Write) into a hardcoded personal Obsidian vault path (/Users/eddale/Documents/COPYobsidian/MAGI/Zettelkasten/...) with no backup step; the skill mitigates this by requiring explicit user approval before edits, but a mistaken approval could overwrite daily-note content.
Comment cette analyse est faite

Exemples

Daily task scan
Scan my tasks for today and tell me if my board is healthy.
Clarify vague todos
Review my daily note and clarify any unclear tasks in Today's 3 and Ready.
Swap today's focus
Swap one task from Ready into Today's 3 and update my daily note.

name: task-clarity-scanner description: Scans daily notes using Personal Kanban structure. Clarifies tasks, manages Today's 3 vs Ready, flags stale items, and helps swap between columns. Use when reviewing todos, scanning task lists, or managing your Kanban board. allowed-tools: Read, Glob, Grep, Edit, Write, AskUserQuestion

Task Clarity Scanner

What This Does

Scans your daily note, identifies unclear or vague tasks, manages the Personal Kanban flow (Today's 3 ↔ Ready), flags stale items, and updates the file once you approve changes.

When to Use

  • "Scan my tasks"
  • "Review my daily note"
  • "Clarify my todos"
  • "Check my task list for today"
  • "What can you help me with today?"
  • "Swap tasks" / "Update my Today's 3"

Default Daily Note Location

Ed's daily notes live in Obsidian at:

/Users/eddale/Documents/COPYobsidian/MAGI/Zettelkasten/YYYY-MM-DD.md

When invoked without a specific file, check today's date and look for that file automatically.

Daily Note Structure (Personal Kanban)

The daily note has this structure:

## Ship This = Win Day
[Single focus - WIP limit 1]

## Today's 3
[Active work - pulled from Ready, WIP limit 3]

## Ready
[Backlog - all carried-forward tasks live here]

## Waiting For
[Blocked/delegated items with dates]

## Done Today
[Completed items]

## Captures
[Links to docs created today]

## Scratch
[Quick notes]

Instructions

This skill uses the Batch Pattern - clarify all tasks first, then execute work.

Note: If you need to triage mobile captures first, use the daily-review agent which runs inbox-triage before this skill.


PASS 0: Kanban Health Check

Before clarifying tasks, assess the board:

  1. Count Today's 3 - Are there exactly 3 tasks? More? Fewer?
  2. Check for stale items - Any tasks marked [STALE]?
  3. Review Ready size - Is the backlog growing out of control?

Report findings:

## Board Status
- Today's 3: [N] tasks (target: 3)
- Ready: [N] tasks
- Stale items: [N] (rolling 3+ days)
- Waiting For: [N] blocked items

If Today's 3 has more than 3 items, offer to help prioritize. If stale items exist, flag them for decision (do, delegate, drop).


PASS 1: Clarify (One by One)

Step 1: Read the Daily Note If no file specified, use today's date to find the daily note in the Zettelkasten folder. Look for tasks in ## Today's 3 and ## Ready sections.

Step 2: Quick Triage Briefly categorize tasks:

  • Clear - Ready to act on
  • Unclear - Needs clarification
  • Stale - Rolling 3+ days, needs decision
  • Done - Already completed, can skip

Step 3: Clarify One at a Time For each unclear task, present it individually (NOT a wall of text):

**Task:** "[the task]"

- **Issue:** [what's unclear]
- **Suggested rewrite:** "[agentic-ready version]"
- **What's needed:** [missing context]

Then ask ONE question with options:

  1. Clarify - "Here's what I mean: [context]"
  2. Accept rewrite - Use the suggested version
  3. Skip - Leave as-is for now
  4. Someday/Maybe - Park it with #someday tag
  5. Create project file - Start a living doc for this task
  6. Move to Ready - Not for today, but keep visible

Move to the next task after each response. Keep momentum.

Step 3a: Stale Item Handling For tasks marked [STALE]:

**Stale Task:** "[STALE] [task text] (MM-DD)"
This has been rolling for [N] days.

Options:
1. **Do it now** - Move to Today's 3, commit to finishing
2. **Delegate** - Move to Waiting For with context
3. **Drop** - Remove entirely (it's not happening)
4. **Reframe** - Break into smaller pieces
5. **Someday** - Park with #someday tag

Step 3b: Project File Creation When user selects "Create project file":

  1. Create in Obsidian Zettelkasten at: /Users/eddale/Documents/COPYobsidian/MAGI/Zettelkasten/PROJECT - [Task Name].md

  2. Seed the file with this template:

---
type: project
status: planning
created: YYYY-MM-DD
linked-from: [[YYYY-MM-DD]]
---
# PROJECT: [Task Name]

## What We're Building
[One paragraph describing the goal and why it matters]

## Constraints & Scope
- In scope: ...
- Out of scope: ...
- Dependencies: ...

## Context Gathered
[Brainstorming notes, research findings, decisions made so far]

## Steps (when ready)
- [ ] Step 1
- [ ] Step 2

## Done State
[How we know this is complete]

## Open Questions
- [ ] Question 1
  1. Update daily note - Replace original task with: - [ ] [[PROJECT - Task Name]] - [brief description]

  2. Offer to continue brainstorming in the project file right now

Step 4: Rewrite Principles When suggesting rewrites:

  • State the specific action
  • Include context needed
  • Define the done state
  • Make it agent-handoff ready
  • Include URLs/links so tasks are self-contained
  • Preserve date suffix - Keep (MM-DD) for staleness tracking

Example:

  • Before: "Make Google Drive AI Ready"
  • After: "Organize Google Drive for AI access: Create 'AI-Ready' folder, move key docs, document what each folder contains (01-04)"

Step 4b: Flag Task Dependencies While clarifying, watch for blocking relationships:

- [ ] Convert project to skill ← Do first
  - [ ] Upload skill ZIP ← Blocked by above

Suggest marking blocked items in Waiting For section.

Step 4c: Surfaced Tasks Clarifying one task often surfaces additional tasks. Track these as you go:

  • New research needed
  • Dependencies discovered
  • Related updates required

These go to the Ready section when updating the file.

Step 4d: Final Check Before moving to PASS 2, ask: "Did we miss anything?"


PASS 1.5: Kanban Swaps (Optional)

If user wants to adjust Today's 3:

Swap Interface:

## Current Today's 3:
1. [Task A]
2. [Task B]
3. [Task C]

## Ready (available to pull):
1. [Task D]
2. [Task E]
3. [Task F]

What would you like to do?
- Swap: "swap 2 with D" - Replace Task B with Task D
- Add: "add E" - Pull Task E into Today's 3 (if under limit)
- Remove: "remove 1" - Move Task A back to Ready
- Done: "done 2" - Mark Task B complete, move to Done Today

Apply changes and confirm.


PASS 2: Update the File

Step 5: Batch the Changes After all tasks are clarified, summarize:

Ready to update your daily note:

## Today's 3 Changes:
- [Task moved in from Ready]
- [Task rewritten]

## Ready Changes:
- Task 1: [original] → [rewrite]
- Task 2: Skipped
- Task 3: → Someday/Maybe (#someday added)

## Stale Items:
- [Task] → Dropped
- [Task] → Moved to Waiting For

## New Tasks (surfaced):
- [New task discovered]

Get final approval before making edits.

Step 6: Apply Edits

  • Use Edit tool to modify the original file
  • Keep tasks in their proper sections (Today's 3 vs Ready)
  • Move completed items to Done Today
  • Move blocked items to Waiting For
  • Add #someday tag for parked items
  • Preserve all other content exactly
  • Confirm: "Updated [N] tasks."

PASS 3: Execute (Future)

Step 7: Spin Up Agents For tasks marked "Ready to Go", offer to spin up agents:

These tasks are ready for agents:
- [task 1] - I can do this now
- [task 2] - Needs a research agent
- [task 3] - I can do this now

Want me to start working on these?

Execute approved tasks, report back when done.

Guidelines

  1. Respect the Kanban structure - Today's 3 = active work, Ready = backlog
  2. Don't over-engineer - A slightly vague task that's obvious to Ed doesn't need rewriting
  3. Preserve voice - Rewrites should sound like Ed, not corporate jargon
  4. Flag unknowns honestly - If you don't know what something means, say so
  5. Suggest, don't assume - Always get approval before modifying
  6. Keep the human in the loop - Help Ed move faster, don't take over
  7. Preserve date suffixes - The (MM-DD) format enables staleness tracking

The Project File Pattern

For tasks that are too big to clarify inline, create a project file instead:

  • The file is a living document that grows over sessions
  • Daily note links to the project file (keeps it visible)
  • When ready to execute, the project file IS the spec
  • If parked for 1+ week, move to [[Someday-Maybe]] file

When to suggest a project file:

  • Task involves multiple sub-tasks
  • Needs research before execution
  • Will take multiple sessions to complete
  • Has dependencies on other work
  • User says "let me think about this"

Research Task Pattern

Research tasks have a distinct completion format:

Format:

- [x] ~~[Research Topic]~~ → **Finding:** [Summary]. See [[Document Name]]

Ultrathink Bundling: When you see 3+ related research tasks, suggest: "These research tasks are related and could run in parallel. Consider bundling as 'Ultrathink - [Topic] Research YYYY-MM-DD'"

Examples

Input (from daily note):

## Today's 3
- [ ] Review PR for auth changes (01-02)
- [ ] [STALE] Fix the login bug (12-30)
- [ ] Write newsletter intro

## Ready
- [ ] Convert Offer Diamond to skill (01-03)
- [ ] Set up Gmail App Password (01-01)
- [ ] Research API options

Output (report):

## Board Status
- Today's 3: 3 tasks (target: 3) ✓
- Ready: 3 tasks
- Stale items: 1 (Fix the login bug - 5 days)
- Waiting For: 0

## Needs Attention

### Stale Item
**Task:** "[STALE] Fix the login bug (12-30)"
Rolling for 5 days. Decision needed:
1. Do it now
2. Delegate
3. Drop
4. Reframe
5. Someday

### Unclear Task
**Task:** "Write newsletter intro"
- Issue: Which newsletter? What's the topic?
- Suggested rewrite: "Write intro for LBR article on [topic] - 2-3 sentences, hook + promise (01-04)"
- What's needed: Article topic and angle

### Ready to Go
- "Review PR for auth changes" - Clear, actionable
- "Convert Offer Diamond to skill" - Clear process exists

Du fichier au premier essai

Le téléchargement contient exactement le contenu affiché sur cette fiche. Cette copie peut différer de la dernière version du dépôt source.

Consulter la source et ses fichiers annexes

Ce téléchargement contient uniquement SKILL.md. Les scripts, références, modèles et dépendances éventuels ne sont pas inclus. Récupérez les fichiers nécessaires depuis la source en conservant leur arborescence avant de poursuivre.

1. Télécharger le fichier

Télécharger SKILL.md

Depuis la racine d’un projet de test, enregistrez le fichier sous .skill-downloads/task-clarity-scanner-mma3jspc/SKILL.md, ou utilisez la commande suivante. Ce dossier sert à examiner la skill avant son installation.

Commande pour macOS/Linux ou WSL : un shell POSIX et curl sont nécessaires. Sous Windows sans WSL, utilisez le bouton de téléchargement et créez le dossier manuellement.

mkdir -p '.skill-downloads/task-clarity-scanner-mma3jspc' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/task-clarity-scanner-mma3jspc/download' --output '.skill-downloads/task-clarity-scanner-mma3jspc/SKILL.md.part' && mv '.skill-downloads/task-clarity-scanner-mma3jspc/SKILL.md.part' '.skill-downloads/task-clarity-scanner-mma3jspc/SKILL.md'

2. Vérifier les prérequis

Lisez SKILL.md avant de l’utiliser. Si le fichier cite des scripts ou services externes, examinez-les et préparez les accès nécessaires dans votre environnement de test.

Outils déclarés par l’auteur (à vérifier dans votre agent) : Read, Glob, Grep, Edit, Write, AskUserQuestion

Aucune section de prérequis reconnue dans ce fichier. Cela ne signifie pas qu’il est autonome : les bibliothèques, outils et accès peuvent être cités ailleurs dans le contenu ou dans le dépôt source.

Les plateformes de la fiche sont des indications du catalogue. Les chemins ci-dessous suivent la documentation des outils ; ils ne prouvent pas que cette skill y a été exécutée. Aucune version minimale de ces outils n’est attestée pour cette fiche.

3. Charger la skill

Après vérification, copiez le dossier complet dans le projet avec l’une des commandes ci-dessous, puis ouvrez une nouvelle session de l’outil. La commande s’arrête si un dossier du même nom existe déjà : comparez les fichiers avant de le remplacer.

Claude Code
test -f '.skill-downloads/task-clarity-scanner-mma3jspc/SKILL.md' && test -s '.skill-downloads/task-clarity-scanner-mma3jspc/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/task-clarity-scanner' && cp -R '.skill-downloads/task-clarity-scanner-mma3jspc' '.claude/skills/task-clarity-scanner'

OpenSkills

OpenSkills doit déjà être installé. La procédure locale install/read est contrôlée avec OpenSkills 1.5.0, qui exige Node.js ≥ 20.6.0. Aucun slug du catalogue n’est résolu automatiquement.

Installation officielle d’OpenSkills
openskills install './.skill-downloads/task-clarity-scanner-mma3jspc'

openskills read 'task-clarity-scanner-mma3jspc'

La lecture doit afficher le contenu du fichier installé et son dossier de base. Elle ne lance ni la skill ni ses dépendances.

Pour un agent qui lit AGENTS.md, synchronisez ensuite l’index. Cette commande modifie AGENTS.md : relisez le diff avant de l’utiliser.

openskills sync

Mode manuel ou autre outil : joignez SKILL.md à une conversation, ou ouvrez-le dans un agent capable de lire les fichiers. Ajoutez les ressources annexes nécessaires. Il n’existe pas de dossier universel reconnu par tous les outils.

4. Faire un premier essai

Utilisez une entrée courte et fictive : un endpoint d’API, un petit document ou un brief d’interface. Dans une nouvelle conversation, joignez le fichier ou donnez son chemin à l’agent, puis adaptez cette demande :

Lis .skill-downloads/task-clarity-scanner-mma3jspc/SKILL.md et applique ses instructions à cet exemple : [décrire une entrée de test]. Indique les fichiers, dépendances ou accès qui manquent. Produis le résultat demandé par la skill sans inventer de données ni exécuter de script non examiné.

Résultat attendu : une sortie correspondant aux instructions de la skill, ou une liste précise des prérequis manquants. Une simple confirmation de lecture ne valide pas le premier usage. Si une ressource externe est indisponible, ce parcours reste à tester.

Skills similaires