todo-triage

Promote pending todo files into approved ready tasks with structured reports.

1|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/lucasmccomb/ccgm --skill todo-triage-lucasmccomb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: todo-triage
Source: https://github.com/lucasmccomb/ccgm/tree/main/modules/todos/skills/todo-triage
Command: npx skills add https://github.com/lucasmccomb/ccgm --skill todo-triage-lucasmccomb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Many review findings are captured as vague pending todos that require human judgement to become actionable work. This Skill walks the pending todo queue, clarifies each item, and promotes scoped, testable Proposed Change paragraphs so resolvers can implement fixes without re-reading the original discussion.

Core Features & Use Cases

  • Interactive triage: Present each pending todo for confirm, skip, modify, or drop decisions and persist the chosen outcome.
  • Modes for automation: Support mode:autofix to auto-promote concrete items and mode:report-only for dry-run classification and recommendations.
  • Safe file operations: Update frontmatter, atomically rename files to keep filename and status in sync, and produce a compact session summary or structured report for programmatic callers.

Quick Start

Run the todo-triage skill to walk .claude/todos, review each pending item, and promote concrete findings to ready with a Proposed change section.

Frequently Asked Questions about todo-triage

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

FAQPage Schema
How do I convert pending todo files into actionable ready tasks?

Todo triage promotes pending todo capture files into scoped, approved ready tasks by applying interactive confirm, skip, modify, or drop decisions to each item and persisting the outcome.

Can I automate triaging markdown todos without manual confirmation?

Yes, todo triage supports mode:autofix to auto-promote concrete items without manual confirmation, and mode:report-only for dry-run classification and recommendations during automated runs.

What is the best way to manage code review findings captured as pending todos?

Todo triage manages code review findings by walking .claude/todos/*.md files, clarifying vague items into scoped Proposed Change paragraphs, and promoting concrete findings to ready status.

Does todo triage work with automated composition from other skills?

Yes, todo triage supports automated composition from other skills by returning a structured report for programmatic callers, updating frontmatter status, and atomically renaming files to keep filename and status in sync.

How do I safely update frontmatter status when promoting todos?

Todo triage safely updates frontmatter status and atomically renames files to keep filename and status in sync during interactive confirm, skip, modify, or drop flows.

When should I use report-only mode instead of interactive todo triage?

Use mode:report-only for dry-run classification and recommendations when you want to review pending todo outcomes before committing changes, rather than running interactive confirm, skip, modify, or drop flows.