check-todos
Audit TODOs against the two-question rubric — eliminate, do-now, or surface to user. No silent TODOs.
- License 许可
- MIT
- 最近更新
- 2026-09-16
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- check-todos
- description
- Audit TODOs against the two-question rubric — eliminate, do-now, or surface to user. No silent TODOs.
03 SKILL.md 正文
Audit TODOs added in the current iteration.
Rubric
Every TODO must answer YES to BOTH:
- Worth doing, now or eventually?
- Requires a design cycle or human review/input before doing?
Explore the code yourself to answer. Outcomes:
- Clear NO to (1) — not worth doing or actively harmful: Delete.
- Clear NO to (2) — doable now without further input: Do it now.
Under uncertainty: lean toward surfacing now for human/product owner review by escalating.
Furthermore: a problem this iteration created or worsened cannot be silently deferred — This must be fixed or escalated for visibility.
Scope
Default: TODOs added or touched in the current diff (uncommitted + commits ahead of upstream/main).
Consult project standards for how to identify TODOs (e.g., TODO.md or other convention).
Procedure
If operating in orchestrator mode, use a subagent to perform the judgement work (e.g., a review-chain:judge). Surface any escalation or uncertainty to the user if needed. Otherwise use an implementation subagent to execute the decision(s), and then normal code review afterward.
04 同仓库更多 Skills
setup-project
Bootstrap a project for the review-chain workflow — set the orchestrator as default agent, add a "Working With Claude Code" section to CLAUDE.md, install the TODO.md + TODO(slug) tracking convention, and optionally wire an auto-format hook. Take it or leave it; the review-chain agents work without any of this.
usage-wall
Ride the 5-hour usage window into the wall — schedule a post-reset wakeup, then resume stalled subagents. No watchdog, no throttling.
usage-guard
Arm a background watchdog on the account's 5-hour usage window
simplify
Review changed code for reuse, quality, efficiency; fix issues found.
configure-models
Generate or edit local per-agent model overrides for review-chain agents (.claude/review-chain-models.conf) — re-pin a model without editing agent files or publishing.
cleanup-editor
Self-edit a draft for clarity, contradictions, answerable open questions. Use after initial draft.