daily-todos

Convert vague work requests into codebase-referenced daily TODO markdown plans.

8|2|Updated May 3, 2026
One-click install
npx skills add https://github.com/ejoongseok/claude-settings --skill daily-todos
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: daily-todos
Source: https://github.com/ejoongseok/claude-settings/tree/main/claude-code/skills/daily-todos
Command: npx skills add https://github.com/ejoongseok/claude-settings --skill daily-todos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you record your day’s tasks and incoming work as a single daily TODO tracker while turning vague requests into actionable, codebase-informed steps instead of generic suggestions.

Core Features & Use Cases

  • Daily single-file TODO tracker: creates or updates exactly one file per day at .local.claude/daily-todos/YYYY-MM-DD.md, organizing “morning plan”, “midday additions”, and “notes”.
  • Codebase-aware task concretization: when you add an item, it references codebase documents (CLAUDE.md, bot/INDEX.md or .local.claude/ONBOARDING.md, .local.claude/modules/*.md, projects, and biz rules) to suggest module-specific entry points, related files, and approach/risks.
  • Auto-priority and task-type tagging: assigns P1/P2/P3 plus tags like [개발], [분석], [회의], [CS], [문서], [인프라], [학습], and updates file content accordingly (including reordering when urgent).
  • Safety guardrails: refuses to guess when the input is too abstract and instead requests clarification, while also flagging situations like an overwhelming TODO backlog.

Quick Start

Ask: “오늘 할일은 고객 CS로 인해 목록 조회가 느려졌다는 원인 분석이랑, 관련 모듈을 코드로 찾아서 개선 접근 방법 제안해줘.”

Frequently Asked Questions about daily-todos

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert rough daily tasks into a codebase-referenced TODO plan?

To convert rough tasks into a codebase-referenced TODO plan, you record incoming work requests so the system can cross-reference repository documentation like module docs and architecture indexes. This process clarifies abstract tasks by suggesting specific entry points, related files, and implementation approaches.

How does codebase-aware task concretization work for developer workflows?

Codebase-aware task concretization works by validating incoming work against existing repository artifacts such as onboarding indexes, module documentation, and business rules. Instead of guessing, it enforces clarification for abstract requests and maps tasks to specific codebase modules and architectural references.

Can I track daily development tasks in a single markdown journal file?

Yes, you can track daily development tasks in a single markdown journaling file created at a deterministic path. The file organizes entries into morning plans, midday additions, and notes, applying auto-priority tagging with P1, P2, or P3 levels and task-type labels like development or analysis.

What is the best way to manage an overwhelming TODO backlog with priority tagging?

The best way to manage an overwhelming TODO backlog is using a daily tracker that applies priority tagging and flags backlog overflow situations. By assigning deterministic priority levels and reordering urgent items, the system ensures critical codebase tasks surface above less pressing daily additions.

Do I need repository documentation to decompose abstract work requests?

Yes, you need repository documentation such as architecture indexes, module docs, and business rules to decompose abstract work requests. The system refuses to guess when input is too vague, enforcing clarification and requiring validated concretization using allowed file reading and searching tools.

Why does task tracking refuse to process vague incoming work requests?

Task tracking refuses to process vague incoming work requests to enforce safety guardrails and deterministic workflow constraints. Without sufficient context from codebase documentation, the system cannot accurately suggest module-specific entry points or approaches, so it requests clarification instead of generating generic suggestions.