How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-todo:todo-addThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
First, resolve the user's home directory by running `echo $HOME` via Bash. The todo file is at `<home>/.claude/todo.md`. The plan directories live under `<home>/.claude/todo/`.
First, resolve the user's home directory by running echo $HOME via Bash. The todo file is at <home>/.claude/todo.md. The plan directories live under <home>/.claude/todo/.
Add a new task to the todo file.
Use $ARGUMENTS as the task description. If no arguments provided, ask the user what task to add.
Every task MUST have a Directory field. If the user did not specify a directory, ask them: "What directory should this task start from?"
To determine the next task number, read the file and find the highest existing ### N. heading, then increment by 1.
Two edits are required:
Add a row to the summary table (in the ## Summary section) with columns: #, Task, Dir, Plan, Status. Set Plan to — and Status to pending.
Append the full task entry under ## Tasks:
### N. Task Title
- **Directory**: `/path/to/directory`
- **Description**:
- Detail lines...
After adding the task, ask the user if they'd like to create an initial plan for it. If yes, create the directory <home>/.claude/todo/<N>/ and a plan.md with the task description and any initial notes.
Confirm what was added.
npx claudepluginhub ccam80/claude-todoCaptures ideas or tasks from conversation context as structured todos with frontmatter, area inference, duplicate detection, and git commits.
Creates tasks or sections in TASKS.md to track project work. Auto-triggers on phrases like 'add a task', 'todo', 'let's plan', or untracked future work.