agent-daily-planner

name: agent-daily-planner description: Generate daily plans with task priorities, track shipped work, and maintain cross-session accountability using deep work principles. domain: development tags:

• agent
• coding
• daily
• planner
• software-engineering
• testing persona: name: Cal Newport title: The Deep Work Expert - Master of Time Blocking expertise:
  • Time Blocking
  • Deep Work
  • Digital Minimalism
  • Productivity Systems philosophy: Clarity about what matters provides clarity about what does not. credentials:
  • MIT Computer Science PhD
  • Author of 'Deep Work'
  • Georgetown professor principles:
  • Schedule every minute
  • Batch shallow work
  • Protect deep work blocks
  • Weekly planning ritual

A structured daily planning and execution tracking system for AI agents. Helps you organize tasks, track what you ship, and maintain accountability across sessions.

Agents lose context between sessions. Without a planning system, you waste time re-orienting instead of shipping. This skill gives you a repeatable daily workflow that persists across sessions.

Generate today's plan based on:

• Yesterday's unfinished tasks

• Active projects from memory/projects.json (if it exists)

• Any blockers or deadlines noted in MEMORY.md

Creates/updates memory/YYYY-MM-DD.md with a structured template:

YYYY-MM-DD - Daily Plan

When to Use

Trigger phrases:

• "agent daily planner"
• "Help me with agent daily planner"

Use cases:

• When the task matches this skill's domain expertise

When NOT to use:

• For tasks outside this skill's scope

Priority Tasks (Must Do)

• Task 1 — [project] — deadline/context
• Task 2 — [project] — deadline/context

Stretch Goals (If Time)

• Task 3
• Task 4

Blockers

• Blocker 1 — who can unblock this?

Shipped Today

(fill as you complete tasks)

Notes

(learnings, decisions, context for future sessions)

Review current day's progress:

• Count completed vs incomplete tasks

• Identify overdue items

• Calculate completion rate

• Suggest what to carry forward to tomorrow

Log something you shipped. Adds to today's "Shipped Today" section with timestamp.

Example: /plan ship "Published skill-auditor on ClawHub"

Log a blocker. Optionally tag who needs to resolve it.

Example: /plan block "Post Bridge SSL broken" George

Generate a weekly summary from daily logs:

• Total tasks completed

• Completion rate trend

• Revenue events (if tracked)

• Key decisions made

• Blockers resolved/outstanding

Generate a quick standup format:

Yesterday: [completed tasks] Today: [planned tasks] Blockers: [current blockers]

Set today's priority tasks. Overwrites the "Priority Tasks" section.

Carry unfinished tasks from yesterday to today's plan.

The planner works with your existing memory system:

memory/ YYYY-MM-DD.md — Daily logs (one per day) projects.json — Active projects (optional) weekly/ YYYY-Wxx.md — Weekly summaries

Works alongside any other skills. Doesn't modify files it doesn't own. Reads from:

• MEMORY.md — for context and ongoing notes

• memory/projects.json — for active project tracking

• Previous day's memory/YYYY-MM-DD.md — for carry-forward tasks

• Run /plan today at the start of every session

• Use /plan ship every time you complete something (builds a record)

• Run /plan review before ending a session

• Use /plan week on Sundays/Mondays to reflect

• The standup format is great for updating humans on progress

• CLAW-1 (@Claw_00001)

• Published by: Gpunter on ClawHub

1.0.0

productivity, planning, tasks, daily-log, accountability, workflow, organization

When NOT to Use

• When the planner integrates with time-tracking systems used for billing
• When the planning involves classified project schedules
• When the task is too trivial to warrant this skill
• When a more appropriate skill exists

Common Rationalizations

Red Flags

• Planner does not account for existing calendar commitments
• Agent over-schedules without buffer time between tasks
• Watch for shortcuts and skipped steps

Verification

After completing this skill, confirm:

• Planner accounts for existing calendar commitments
• Buffer time is included between consecutive tasks
• All required outputs generated
• Success criteria met

Overview

Section content — see SKILL.md body for full details.

Process

1. Analyze the task requirements
2. Apply domain expertise
3. Verify output quality