メニュー
Create and revise a structured topic outline from an evolving brief, audience, angle, thesis, sections, questions, and sources.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
SOC 職業分類に基づく
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/xiaotianfotos/homerail --skill topic-outlineコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
| name | topic-outline |
| description | Create and revise a structured topic outline from an evolving brief, audience, angle, thesis, sections, questions, and sources. |
Use this Skill when the user asks to plan, structure, or revise a content topic. Do not use it for a generic task checklist or for an outline the user did not request.
Call the current Tool catalog entry for com.homerail.topic-outline:upsert_topic_outline with the complete current outline. Its harness-safe wire name is supplied by the current turn and must not be guessed. Reuse one stable plugin-owned id such as com.homerail.topic-outline:current across turns so a revision replaces the same semantic node without colliding with Core or another plugin. A partial outline is valid: record what is known, preserve useful prior fields, and put only unresolved decisions in questions.
Keep the shape concise:
title names the topic, not the UI component.brief, audience, angle, and thesis capture the current editorial direction.outline contains ordered sections with short points and truthful status.sources includes only real HTTP(S) references supplied or found through an authorized tool.next_action states the next useful editorial step.confidence reflects evidence, not presentation polish.Use the matching topic-outline Tool only when it is listed for the current turn. Do not substitute a legacy dynamic widget. If the requested outline is not placed on the canvas, tell the user only that visible result and offer to retry.
Run HomeRail's built-in read-only pull request review DAG. Use when the user asks to review a GitHub PR, requests an evidence-backed PR audit, or wants a reusable independent Qwen, Kimi, and GLM review with retained findings.
Run, monitor, inspect, debug, and author HomeRail DAG workflows through the CLI or Manager Tools, including supervised multi-Actor runs with stable live Surfaces and later per-Actor follow-ups. Use when: (1) listing templates, (2) starting a DAG run, (3) supervising or watching a run, (4) inspecting chats, handoffs, scorecard, eval-run, or replay output, (5) creating multiple parallel UI panels that update while Actors work, (6) sending commands to waiting Actors, (7) injecting instructions into running nodes, or (8) creating custom DAG templates. For installing or starting Manager/Node/Worker services, use homerail-install-ops first.
Select, inspect, instantiate, adapt, and validate HomeRail's built-in DAG design patterns. Use when an agent needs to design a reusable workflow for periodic triage, read-only issue diagnosis, planner/worker fan-out, on-demand executor/advisor consultation, budget admission, graduated trust, standing-goal verification, quorum decisions, adversarial builder/breaker review, monotonic improvement, or evidence-driven process evolution. Also use when deciding whether a workflow needs condition, loop, join, or while gateway semantics. Patterns describe control-flow invariants; for stable live Actor panels, Surface updates, or follow-up commands, use homerail-dag-ops and a concrete presentation-aware Workflow.