Litho is a Rust-based AI documentation generator that analyzes a codebase and produces architecture documentation in the C4 model format. Developers, architects, and technical leads use it to create and maintain diagrams and code-level documentation, while the catalogue provides skills for working with the project.
Borrowing it
Nothing to install: this file belongs to sopaco/deepwiki-rs. 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/sopaco/deepwiki-rs/main/.agents/skills/litho-documents-skill/SKILL.mdgit clone --depth 1 https://github.com/sopaco/deepwiki-rsWrote 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/sopaco/deepwiki-rs/litho-documents-skill)<a href="https://agentmods.dev/skills/sopaco/deepwiki-rs/litho-documents-skill"><img src="https://agentmods.dev/badge/skills/sopaco/deepwiki-rs/litho-documents-skill.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00185 | $0.02238 |
| Opus 5 | $0.00093 | $0.01119 |
| Sonnet 5 | $0.00037 | $0.00448 |
| Haiku 4.5 | $0.00018 | $0.00224 |
Grade A, and why
litho-document-skill 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 7d 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Litho Document Skill（纯 Agent 版）
本 Skill 是 Litho（deepwiki-rs）的纯 Agent 平行实现。不依赖任何外部二进制，完全通过 Agent 的工具调用能力自主完成四阶段文档生成流水线。
目标产出：
1.概述.md— C4 Context 图 + 项目概述 + 业务价值2.架构.md— C4 Container/Component 图 + 架构模式 + 模块职责3.工作流.md— 时序图 + 流程图 + 并发模型 + 错误处理4.Deep-Exploration/— 每个领域模块的深度研究文档5.边界接口.md— CLI/API/配置等对外接口清单6.数据库概览.md— ER 图 + 表结构（条件触发）
四阶段流水线总览
预处理 → 研究 → 编排 → 输出
↓ ↓ ↓ ↓
结构洞察 C1-C4 Markdown 文件持久化
每个阶段的详细执行指南在 references/ 中，Agent 按需加载。下面只给出决策级指导。
阶段一：预处理 → 了解项目
决策要点：
- 根据项目规模选择扫描策略（见下方快速路径）
- 建立预处理报告：项目名、语言、框架、核心模块列表、README摘要
- 预处理报告是后续所有阶段的基础上下文，务必准确
快速路径（按项目规模）：
| 规模 | 判断标准 | 扫描策略 |
|---|---|---|
| 小 | <100 源文件 | list_files 递归 + read_file 全部核心文件 |
| 中 | 100-500 源文件 | list_files 仅一级目录 + read_file 入口+配置+README + codebase_search 语义搜索 |
| 大 | >500 源文件 | 仅读 README + 主配置 + 入口文件 + view_file_outline 核心模块 + grep_search 精确搜索 |
详细步骤见
references/phase1-preprocessing.md
阶段二：研究 → C4 多层级分析
决策要点：
- 执行顺序：C1 → C2 → [C3 并行]（与 deepwiki-rs 一致）
- 领域模块必须全覆盖：
src/下每个子目录都识别为候选模块，用 DDD 分组（核心域/支撑域/通用域），不得遗漏 - 渐进式深度控制：按 importance 评分分级分析
- 研究产出写入
.litho-agent/临时目录持久化（见下方中间产物策略）
并发搜索：Step 2.3(架构) + 2.4(工作流) + 2.6(边界) 的搜索可并发调用，Step 2.5(模块深度) 必须在 2.2(领域模块) 之后
渐进式深度：
| importance | 分析深度 | 读取文件数 | Mermaid 图 |
|---|---|---|---|
| ≥7（核心域） | 深度分析 | 5+ | 完整 flowchart + 交互表格 |
| 4-6（支撑域） | 标准分析 | 3 | 精简流程图 |
| ≤3（通用域） | 简要描述 | 1-2 | 无图 |
详细步骤见
references/phase2-research.md
阶段三：编排 → 生成 Markdown 文档
决策要点：
- 生成顺序：边界接口 → 概述 → 模块深度(逐个) → 架构 → 工作流 → 数据库（依赖少的先写入）
- 分章节写入大型文档：架构和工作流分 2-3 次写入（框架 → 补充章节）
- 逐模块独立写入：每个 Deep-Exploration 文档独立 write_to_file，写完即释放上下文
- 代码引用密度：每模块 ≥3 文件路径、≥2 类型名、组件表每行有路径列
⚠️ 叙述性写作风格（P0 关键！）：
生成的文档必须面向人类阅读友好，而不是冷冰冰的 PPT 式结构化文字。核心要求：
- 每个章节开头必须有叙述性 summary：用 2-4 句话先解释"这个章节在说什么、为什么重要"，不要直接甩出表格或列表
- 表格和列表前后必须有解读段落：不要只给结构化数据，要解释"这意味着什么、为什么这样设计"
- 设计决策必须讲"为什么"：不只说"选了什么"，还要说"放弃了什么、为什么这样选"
- 用类比和比喻建立理解桥梁：比如把 Memory 比作"快递站"、把 Agent 比作"工人"、把 Pipeline 比作"生产线"
- 避免冷冰冰的标题堆叠：章节标题应该自然引出叙述，而不是
### 2.1 核心目标→ 直接跳到列表
What ships with it
5 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.
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.
- 7d ago First seen · 161 lines · 185 tokens per session scan A fd12f6cec763
litho-document-skill is a skill published in the GitHub repository sopaco/deepwiki-rs (1,714 stars, last pushed 23d ago), licensed MIT. It adds 185 tokens to every session and 2,238 once invoked, about $0.0009 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
skill-creator
Create, install, or update skills in the workspace. Use when (1) installing a skill from a URL or remote source, (2) creating a new skill from scratch, (3) updating or restructuring existing skills. Always use this skill for any skill installation or creation task.
image-generation
Generate or edit images from text prompts. Use when the user asks to create, draw, design, or edit an image, illustration, photo, icon, poster, or any visual content.
knowledge-wiki
Manage the personal knowledge wiki. Use when the user shares articles, documents, or asks to organize knowledge; when a conversation produces insights worth preserving as structured knowledge; or when the user asks about the knowledge base.
screen-ocr-operator
Use when you need to see and operate an Android screen through an OCR/vision model plus ADB. Optimized commander workflow: one model plan, batch ADB execution, minimal round-trips, verify at milestones.
device-shell
Use when you need to execute shell commands on an Android device from a Linux/proot environment. Covers the ADB channel and an optional local Shizuku HTTP shell bridge.
csswitch-external-skill-tools
A routing guide for installing or removing external skills in CSSwitch-managed Claude Science environments. An external skill is an add-on brought in from another source.