Run any Skill in Manus with one click

analyze-reference

Stars0

Forks0

UpdatedApril 27, 2026 at 02:35

拆解参考作品、提炼节奏、情绪、场景结构与禁用模式。Use when benchmarking imported novels from projects/<project-id>/references/raw/ and writing v2 outputs into source_assets/reference_analysis/.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Run Skill in Manus

Source

luoluomashang

luoluomashang/novel-engine

View GitHub Repository View Creator Repositories

Download

Run Skill in Manus

Related occupationsSOC

Based on SOC occupation classification

Writers and AuthorsArts, Design, Entertainment, Sports, and Media Occupations·SOC 27-3043

SKILL.md

readonly

name	analyze-reference
description	拆解参考作品、提炼节奏、情绪、场景结构与禁用模式。Use when benchmarking imported novels from projects/<project-id>/references/raw/ and writing v2 outputs into source_assets/reference_analysis/.
argument-hint	<project-id>
disable-model-invocation	true

拆解参考书

参数

$0：project_id

执行步骤

若 $0 缺失，立即停止并要求用户提供 project_id。
先运行 python app.py ensure-prereq "$0" reference_analysis。
调用 style-miner 子代理执行真实对标分析。
style-miner 必须直接写入：
- projects/$0/source_assets/reference_analysis/reference_manifest.json
- projects/$0/source_assets/reference_analysis/reference_summary.json
- projects/$0/source_assets/reference_analysis/reports/<reference>.json
长文本至少覆盖前段 / 中段 / 后段；先剔除站点声明、上传说明、版权前缀。
不要运行 python app.py analyze-reference "$0" 或 scripts/analyze_reference.py 来生成内容；这些脚本只保留给测试与回归。
返回中文摘要，并建议进入 /build-dna。

输出要求

source_assets/reference_analysis/reports/*.json 中的文本说明统一中文。
结论必须能支撑后续 DNA 编译，而不是只给宽泛评价。

阶段产物契约

输入文件：projects/$0/references/raw/*、已有的 projects/$0/reference_info.json。
输出文件：projects/$0/source_assets/reference_analysis/reference_manifest.json、projects/$0/source_assets/reference_analysis/reference_summary.json、projects/$0/source_assets/reference_analysis/reports/*.json。
禁止行为：不要运行 python app.py analyze-reference 或 scripts/analyze_reference.py 代写内容；不要把产物写回旧版 analysis/ 或 references/parsed/ 目录；不要只读开头几段就下结论。

More from this repository

same repository

authoring-room

luoluomashang/novel-engine

生成 authoring_setup proposal，并通过 scoped publish 更新 project_playbook、input_policy、output_policy、failure_mode_policy 或 planning_branch_modes。Use when authoring contract needs setup or revision.

2026-04-270

booming

luoluomashang/novel-engine

生成高烈度改局 planning branch proposal，只产出分支提案，不直接写正式 chapter planning 产物。

2026-04-270

build-dna

luoluomashang/novel-engine

根据参考拆解结果编译 style_dna、scene_formulas、reader_hook_map、banned_patterns 和 prose_constraints。Use when source_assets/reference_analysis/reference_summary.json and reports/ already exist.

2026-04-270

chapter-room

luoluomashang/novel-engine

根据 published views 与 chapter planning source 生成章节 proposal；必要时调用 discussion agents，并通过 confirm-stage chapter_planning 生效。Use when chapter_planning is ready or a chapter plan needs revision.

2026-04-270

character-room

luoluomashang/novel-engine

为单张角色卡生成或修订 proposal，并通过 scoped confirm-stage materialize。Use when a character card needs creation, revision, lifecycle change, or reactivation.

2026-04-270

compile-context

luoluomashang/novel-engine

检索 effective views 与章节归档，并生成当前章节的 published compiled packs。Use when scene beats, context_permission_source, and context_query are ready.

2026-04-270

name	analyze-reference
description	拆解参考作品、提炼节奏、情绪、场景结构与禁用模式。Use when benchmarking imported novels from projects/<project-id>/references/raw/ and writing v2 outputs into source_assets/reference_analysis/.
argument-hint	<project-id>
disable-model-invocation	true

拆解参考书

参数

$0：project_id

执行步骤

若 $0 缺失，立即停止并要求用户提供 project_id。
先运行 python app.py ensure-prereq "$0" reference_analysis。
调用 style-miner 子代理执行真实对标分析。
style-miner 必须直接写入：
- projects/$0/source_assets/reference_analysis/reference_manifest.json
- projects/$0/source_assets/reference_analysis/reference_summary.json
- projects/$0/source_assets/reference_analysis/reports/<reference>.json
长文本至少覆盖前段 / 中段 / 后段；先剔除站点声明、上传说明、版权前缀。
不要运行 python app.py analyze-reference "$0" 或 scripts/analyze_reference.py 来生成内容；这些脚本只保留给测试与回归。
返回中文摘要，并建议进入 /build-dna。

输出要求

source_assets/reference_analysis/reports/*.json 中的文本说明统一中文。
结论必须能支撑后续 DNA 编译，而不是只给宽泛评价。

阶段产物契约

输入文件：projects/$0/references/raw/*、已有的 projects/$0/reference_info.json。
输出文件：projects/$0/source_assets/reference_analysis/reference_manifest.json、projects/$0/source_assets/reference_analysis/reference_summary.json、projects/$0/source_assets/reference_analysis/reports/*.json。
禁止行为：不要运行 python app.py analyze-reference 或 scripts/analyze_reference.py 代写内容；不要把产物写回旧版 analysis/ 或 references/parsed/ 目录；不要只读开头几段就下结论。