docs-engineering — for Claude Code docs-engineering, ShioRamen, community, for Claude Code, ide skills, coding-assistant, llamacpp, self-hosted, **Public API surface** — read, shio.toml

v1.0.0

关于此技能

适用场景: Ideal for AI agents that need step 1: discover documents. 本地化技能摘要: A local AI coding assistant powered by llama.cpp. It covers coding-assistant, llamacpp, llm workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

功能特性

Step 1: Discover documents
Find all documentation files in the project (excluding vendored deps):
find . -name " .md" ! -path "./vendor/ " ! -path "./target/ "
Also check config and schema files that double as documentation:
find . -name " .toml" ! -path "./vendor/ " ! -path "./target/ " -not -name "Cargo.lock"

# 核心主题

hiroshiyui hiroshiyui
[0]
[0]
更新于: 4/8/2026

Killer-Skills Review

Decision support comes first. Repository text comes second.

Reference-Only Page Review Score: 10/11

This page remains useful for teams, but Killer-Skills treats it as reference material instead of a primary organic landing page.

Original recommendation layer Concrete use-case guidance Explicit limitations and caution Quality floor passed for review
Review Score
10/11
Quality Score
67
Canonical Locale
en
Detected Body Locale
en

适用场景: Ideal for AI agents that need step 1: discover documents. 本地化技能摘要: A local AI coding assistant powered by llama.cpp. It covers coding-assistant, llamacpp, llm workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

核心价值

推荐说明: docs-engineering helps agents step 1: discover documents. A local AI coding assistant powered by llama.cpp. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

适用 Agent 类型

适用场景: Ideal for AI agents that need step 1: discover documents.

赋予的主要能力 · docs-engineering

适用任务: Applying Step 1: Discover documents
适用任务: Applying Find all documentation files in the project (excluding vendored deps):
适用任务: Applying find . -name " .md" ! -path "./vendor/ " ! -path "./target/ "

! 使用限制与门槛

  • 限制说明: Read the actual source files; do not guess from the docs what the code does.
  • 限制说明: Keep the existing prose style and structure; only change what is incorrect.
  • 限制说明: Do not reformat sections you aren't fixing.

Why this page is reference-only

  • - Current locale does not satisfy the locale-governance contract.

Source Boundary

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

评审后的下一步

先决定动作,再继续看上游仓库材料

Killer-Skills 的主价值不应该停在“帮你打开仓库说明”,而是先帮你判断这项技能是否值得安装、是否应该回到可信集合复核,以及是否已经进入工作流落地阶段。

实验室 Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

常见问题与安装步骤

以下问题与步骤与页面结构化数据保持一致,便于搜索引擎理解页面内容。

? FAQ

docs-engineering 是什么?

适用场景: Ideal for AI agents that need step 1: discover documents. 本地化技能摘要: A local AI coding assistant powered by llama.cpp. It covers coding-assistant, llamacpp, llm workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

如何安装 docs-engineering?

运行命令:npx killer-skills add hiroshiyui/ShioRamen/docs-engineering。支持 Cursor、Windsurf、VS Code、Claude Code 等 19+ IDE/Agent。

docs-engineering 适用于哪些场景?

典型场景包括:适用任务: Applying Step 1: Discover documents、适用任务: Applying Find all documentation files in the project (excluding vendored deps):、适用任务: Applying find . -name " .md" ! -path "./vendor/ " ! -path "./target/ "。

docs-engineering 支持哪些 IDE 或 Agent?

该技能兼容 Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer。可使用 Killer-Skills CLI 一条命令通用安装。

docs-engineering 有哪些限制?

限制说明: Read the actual source files; do not guess from the docs what the code does.;限制说明: Keep the existing prose style and structure; only change what is incorrect.;限制说明: Do not reformat sections you aren't fixing.。

安装步骤

  1. 1. 打开终端

    在你的项目目录中打开终端或命令行。

  2. 2. 执行安装命令

    运行:npx killer-skills add hiroshiyui/ShioRamen/docs-engineering。CLI 会自动识别 IDE 或 AI Agent 并完成配置。

  3. 3. 开始使用技能

    docs-engineering 已启用,可立即在当前项目中调用。

! 参考页模式

此页面仍可作为安装与查阅参考,但 Killer-Skills 不再把它视为主要可索引落地页。请优先阅读上方评审结论,再决定是否继续查看上游仓库说明。

Upstream Repository Material

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

Upstream Source

docs-engineering

A local AI coding assistant powered by llama.cpp. It covers coding-assistant, llamacpp, llm workflows. This AI agent skill supports Claude Code, Cursor, and

SKILL.md
Readonly
Upstream Repository Material
The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.
Supporting Evidence

Docs Engineering

Audit and repair project documentation so it accurately reflects the current state of the code. This skill covers validity (broken links, stale examples), sync (docs match the implementation), and completeness (new features are documented).

Step 1: Discover documents

Find all documentation files in the project (excluding vendored deps):

bash
1find . -name "*.md" ! -path "./vendor/*" ! -path "./target/*"

Also check config and schema files that double as documentation:

bash
1find . -name "*.toml" ! -path "./vendor/*" ! -path "./target/*" -not -name "Cargo.lock"

Step 2: Understand the code surface

Before auditing docs, understand what the code actually exposes. For this Rust project:

Commands and CLI flags — read the clap struct definitions:

bash
1grep -r "struct.*Args\|#\[command\]\|#\[arg\]" src/ --include="*.rs" -l

Config keys — read the config structs:

bash
1grep -r "struct.*Config\|#\[serde" src/ --include="*.rs" -A 3

Public API surface — read src/main.rs and submodules.

Read the actual source files; do not guess from the docs what the code does.

Step 3: Audit each document

For each document, check:

3a. CLI option tables

Compare every flag listed in the docs against the actual clap definitions in source. For each discrepancy:

  • Flag listed in docs but not in code → remove it
  • Flag in code but missing from docs → add it
  • Description or default value differs → correct it

3b. Config file reference

Compare every key in the shio.toml reference block against the config structs in src/. For each discrepancy:

  • Key listed in docs but not in struct → remove it
  • Field in struct missing from docs → add it with its type and default
  • Wrong type or default → correct it

3c. Code examples

For each shell command block in docs, verify it would actually work:

  • Binary names match what cargo install produces
  • Flags match current CLI
  • Paths like ./bin/llama-server are consistent throughout

Check all relative markdown links ([text](path)):

bash
1grep -o '\[.*\]([^)]*\.md[^)]*)' README.md

Verify each linked file exists. Fix or remove broken links.

3e. External URLs

Spot-check external links (GitHub repos, official docs). Flag any that look stale or wrong — do not auto-fix external URLs without confirming with the user.

Step 4: Repair

Make the minimum edits needed to make docs accurate:

  • Edit the document to match the code (not the other way around) unless the code itself appears to be wrong.
  • Keep the existing prose style and structure; only change what is incorrect.
  • Do not reformat sections you aren't fixing.
  • Do not add documentation for features not yet implemented.

Step 5: Maintain TODO.md

For any doc/TODO.md (or equivalent task-tracking document) in the project:

5a. Mark completed tasks

Cross-reference each task/step against the actual code on disk and git history. A task is done if:

  • The files it describes exist and contain the expected content, or
  • git log shows a commit that implemented it

Mark done tasks by prefixing the heading with [x]:

markdown
1### [x] A1 — Add mRuby git submodule

5b. Summarize completed phases

When all steps in a phase are done, replace the entire phase section (which may contain multi-screen code listings and prose) with a compact summary block:

markdown
1## Phase A — Infrastructure ✓ DONE 2 3**Files created:** `build.rs`, `mruby_configs/shio.rb`, `mruby_configs/mcp_safe.gembox`, 4`src/ruby/{ffi.rs, glue.c, native.rs, prelude.rb, vm.rs, registry.rs, mod.rs}` 5 6**Key decisions recorded:** 7- mRuby pinned to commit `a309524d0` (same as rrcad) 8- gembox: `stdlib` + `math` + `mruby-compiler` only (security boundary) 9- `#![allow(dead_code)]` on Phase A stubs — removed in Phase B when used 10 11**Verified:** `cargo build` ✓ · 320 tests pass ✓ · `clippy -D warnings` ✓ · `cargo fmt`

Keep the summary tight: one sentence per key decision, one line for files created, one line for verification status. Drop all code listings — they're in git.

5c. Remove completed items

Once a task or phase has been fully summarized (per 5b), delete the completed entries from doc/TODO.md to keep the file focused on remaining work. If a completed item contains useful reference information (key decisions, gotchas, API notes), move that content to doc/reference_manual.md before removing it from TODO.md.

5d. Leave pending phases untouched

Do not modify any phase that is not yet started or only partially done. Only summarize when the entire phase (every numbered step) is verifiably complete.

Step 6: Verify the example config

The shio.toml in the repo root is an example config. Cross-check it against src/config.rs (or equivalent): every key present in the file must be a recognized field. Update the example file if needed.

5a. Sync system_promptDEFAULT_SYSTEM_PROMPT

shio.toml [chat].system_prompt must be kept word-for-word in sync with DEFAULT_SYSTEM_PROMPT in src/chat.rs. They serve the same role: shio.toml is the editable copy users customise; the Rust constant is the compile-time fallback used when no config file is present.

  • Read DEFAULT_SYSTEM_PROMPT from src/chat.rs.
  • Read system_prompt from shio.toml.
  • If they differ, update shio.toml to match src/chat.rs (the Rust source is authoritative).
  • TOML """ strings use \ as a line-ending escape (same as Rust), so the formatting translates directly.

Step 6: Report

After completing repairs, give the user a concise summary:

Docs audit complete.

Fixed:
- README.md: removed --flash-attn flag (renamed to --flash-attention in v0.x)
- README.md: added --system-prompt option to `serve` and `chat` sections
- shio.toml: corrected cache_type_k default from "q4_0" to "f16"

No issues:
- Internal links: all valid
- Command examples: all match current CLI

Skipped (needs confirmation):
- [External URL] https://... — looks potentially stale, please verify

Guidelines

Do:

  • Read source before editing docs — trust the code, not memory
  • Make targeted edits; preserve structure and voice
  • Check both the narrative docs AND the inline code/config examples

Do not:

  • Rewrite docs wholesale when a small correction suffices
  • Document aspirational or planned features as if they exist
  • Change code to match docs without flagging it to the user first
  • Touch vendor/ documentation

相关技能

寻找 docs-engineering 的替代方案 (Alternative) 或可搭配使用的同类 community Skill?探索以下相关开源技能。

查看全部

openclaw-release-maintainer

Logo of openclaw
openclaw

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

333.8k
0
AI

widget-generator

Logo of f
f

为prompts.chat的信息反馈系统生成可定制的插件小部件

149.6k
0
AI

flags

Logo of vercel
vercel

React 框架

138.4k
0
浏览器

pr-review

Logo of pytorch
pytorch

Python中具有强大GPU加速的张量和动态神经网络

98.6k
0
开发者工具