Borrowing it
Nothing to install: this file belongs to dama-cyber/magic-distillation. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/dama-cyber/magic-distillation/main/.claude/skills/fiction-distiller/SKILL.mdgit clone --depth 1 https://github.com/dama-cyber/magic-distillationWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/dama-cyber/magic-distillation/fiction-distiller)<a href="https://agentmods.dev/skills/dama-cyber/magic-distillation/fiction-distiller"><img src="https://agentmods.dev/badge/skills/dama-cyber/magic-distillation/fiction-distiller/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/dama-cyber/magic-distillation/fiction-distiller"><img src="https://agentmods.dev/badge/skills/dama-cyber/magic-distillation/fiction-distiller.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00064 | $0.04013 |
| Opus 5 | $0.00032 | $0.02006 |
| Sonnet 5 | $0.00013 | $0.00803 |
| Haiku 4.5 | $0.00006 | $0.00401 |
Grade A, and why
fiction-distiller scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 244 lines — stays where its author put it; the contents beside it link to each section on GitHub.
fiction-distiller
短篇换元仿写技能 | 版本 4.1（核心修复版）
知识路径解析
本技能中 @skill:knowledge/... 引用指向 SKILL.md 同级的 knowledge/ 目录。映射规则：
@skill:knowledge/index.md → knowledge/index.md
@skill:knowledge/general-style/01-stylistics → knowledge/general-style/01-stylistics.md
所有 @skill:knowledge/X/Y 均对应 knowledge/X/Y.md（自动追加 .md 扩展名）。加载时直接读取对应路径。
输出路径
所有生成的内容统一保存到 SKILL.md 同级的 output/ 目录下。按类型和章节逐文件保存，一章一文件。
严禁修改或覆盖 output/ 以外的任何项目文件（SKILL.md、knowledge/、CLAUDE.md、README.md 等）。只读读取，只写 output。
| 内容类型 | 命名规则 | 示例 |
|---|---|---|
| 风格模型卡 | 风格模型卡-《作品名》.md |
风格模型卡-《百年孤独》.md |
| 长篇正文 | 第N章-《故事名》.md |
第01章-《雾港》.md |
| 长篇大纲 | 大纲-《故事名》.md |
大纲-《雾港》.md |
| 短篇正文 | 《故事名》.md |
《回声》.md |
| 分析报告 | 分析-《作品名》.md |
分析-《局外人》.md |
角色定位
你是世界顶尖的文学风格分析师与创意写作大师。你拥有深厚的文学理论功底、敏锐的风格感知力和精湛的创作技艺。你的核心能力是：从任何小说文本中蒸馏出风格精华，并以该风格创作全新的原创故事。
懒加载协议
禁止一次性加载全部知识文件。严格按三层协议分步加载，每一步完成后再触发生下一步：
[触发条件] [加载内容] [完成后]
步骤 1（知识索引） → index.md 摘要 → 触发步骤 2
步骤 2（文本接收后） → 基础层（始终加载） → 触发步骤 3
步骤 3（蒸馏开始） → 体裁层（按类型加载） → 触发步骤 4-5
步骤 4-5（创作阶段） → 方法层（按需加载） → 触发步骤 6
| 步骤 | 触发时机 | 加载范围 | 文件 |
|---|---|---|---|
| ① 索引 | 始终首先执行 | 仅 index.md 摘要 | @skill:knowledge/index.md |
| ② 基础层 | 步骤 2 文本接收后 | 通用分析 + 指纹方法论（11 个文件） | general-style/* + author-fingerprint/* |
| ③ 体裁层 | 步骤 3 蒸馏开始时 | 按文本类型选一组（5-7 个文件） | long-fiction/* / short-fiction/* / net-style/* |
| ④ 方法层 | 步骤 4-5 创作阶段 | 创作方法 + 验证（4 个文件） | ai-distillation/* |
全自动执行流程
步骤 1：知识索引构建
读取 @skill:knowledge/index.md，按三层优先级构建分析矩阵：
- 基础层（始终首先加载摘要）：
general-style/*+author-fingerprint/* - 体裁层（按用户文本类型选择性深入加载）：
long-fiction/*/short-fiction/*/net-style/* - 方法层（创作阶段加载）：
ai-distillation/*
步骤 2：文本接收与智能切分
| 文本类型 | 自动处理方式 |
|---|---|
| 长篇（>5万字） | 按章节标题/分卷标记/逻辑节点切分 → 逐章独立蒸馏 → 汇总跨章节规律 → 生成全书风格模型卡 |
| 短篇（<5万字） | 直接全文蒸馏 |
| 多篇文本 | 逐篇蒸馏 → 对比分析 → 提取跨作品风格核心 |
步骤 3：全维度风格蒸馏
从以下 11 个维度逐一解构，汇总为风格模型卡：
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- knowledge/ai-distillation/01-feature-extraction.md 8.2 KB
- knowledge/ai-distillation/02-fingerprint-verification.md 7.2 KB
- knowledge/ai-distillation/03-imitation-strategies.md 9.5 KB
- knowledge/ai-distillation/04-evaluation.md 8.5 KB
- knowledge/anti-detection-short/01-detection-principles.md 5.9 KB
- knowledge/anti-detection-short/02-humanization.md 6.2 KB
- knowledge/anti-detection-short/03-rhythm-perturbation.md 5.3 KB
- knowledge/anti-detection-short/04-ai-word-list.md 5.7 KB
- knowledge/anti-detection-short/05-short-verification.md 5.7 KB
- knowledge/author-fingerprint/01-lexical.md 8.8 KB
- knowledge/author-fingerprint/02-syntactic.md 9.1 KB
- knowledge/author-fingerprint/03-rhetorical.md 8.6 KB
- knowledge/author-fingerprint/04-narrative-voice.md 8.9 KB
- knowledge/author-fingerprint/05-punctuation.md 8.3 KB
- knowledge/author-fingerprint/06-cross-work.md 8.0 KB
- knowledge/general-style/01-stylistics.md 8.6 KB
- knowledge/general-style/02-narratology.md 7.8 KB
- knowledge/general-style/03-rhetoric.md 8.7 KB
- knowledge/general-style/04-stylometry.md 8.5 KB
- knowledge/general-style/05-imaging.md 8.9 KB
- knowledge/index.md 9.9 KB
- knowledge/long-fiction/01-story-layer.md 8.0 KB
- knowledge/long-fiction/02-narrative-layer.md 7.8 KB
- knowledge/long-fiction/03-structure.md 7.9 KB
- knowledge/long-fiction/04-style.md 7.6 KB
- knowledge/long-fiction/05-revision.md 7.7 KB
- knowledge/net-style/01-genres.md 11 KB
- knowledge/net-style/02-pleasure-mechanism.md 11 KB
- knowledge/net-style/03-world-building.md 9.6 KB
- knowledge/net-style/04-characters.md 11 KB
- knowledge/net-style/05-rhythm-structure.md 8.7 KB
- knowledge/net-style/06-narrative-focus.md 9.5 KB
- knowledge/net-style/07-era-features.md 9.0 KB
- knowledge/originality-check/01-ngram-check.md 4.5 KB
- knowledge/originality-check/02-variable-residue.md 5.9 KB
- knowledge/originality-check/03-human-review.md 5.1 KB
- knowledge/short-fiction/01-elements.md 7.9 KB
- knowledge/short-fiction/02-strategic-choice.md 7.7 KB
- knowledge/short-fiction/03-details.md 7.4 KB
- knowledge/short-fiction/04-structure.md 7.7 KB
- knowledge/short-fiction/05-three-beauties.md 8.5 KB
- knowledge/short-substitution/01-what-is-substitution.md 4.1 KB
- knowledge/short-substitution/02-skeleton-extraction.md 6.5 KB
- knowledge/short-substitution/03-variable-list.md 7.4 KB
- knowledge/short-substitution/04-substitution-levels.md 7.3 KB
- knowledge/short-substitution/05-word-count-budget.md 6.0 KB
- knowledge/short-substitution/06-short-pipeline.md 7.4 KB
- knowledge/word-count-control/01-count-analysis.md 4.0 KB
- knowledge/word-count-control/02-budget-table.md 4.5 KB
- knowledge/word-count-control/03-expansion-compression.md 5.9 KB
- output/.gitkeep 4 B
- prompts/p01-skeleton-extract.md 3.8 KB
- prompts/p02-variable-annotate.md 4.9 KB
- prompts/p03-new-variable-design.md 5.3 KB
- prompts/p04-paragraph-generate.md 4.9 KB
- prompts/p05-humanize.md 3.8 KB
- prompts/p06-word-count-fix.md 3.2 KB
- prompts/p07-self-check.md 4.6 KB
- tools/anti_detection_checker.py 22 KB runs code
- tools/batch_extract.py 17 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 244 lines · 64 tokens per session scan A b31bed1d5c25
fiction-distiller is a skill published in the GitHub repository dama-cyber/magic-distillation (100 stars, last pushed 2mo ago), licensed MIT. It adds 64 tokens to every session and 4,013 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…