Token导航 LogoToken导航

Agent Skill · 研究检索

code-tour(代码之旅)

用于围绕 GitHub 仓库、Issue、Pull Request、分支、提交和代码协作流程提供辅助能力。它适合让 Agent 查询项目状态、整理变更、辅助创建或检查协作事项,并把仓库中的信息转成可执行的下一步。使用时需要区分只读查询和写入操作;涉及创建 PR、修改 Issue、推送分支或访问私有仓库时,应确认 token 权限、目标仓库范围和用户授权。

code-tour 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 12,853。

总安装 9,914周安装 405GitHub Stars 31,737下载 3,208核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:code-tour(代码之旅)
来源仓库:https://github.com/github/awesome-copilot
仓库路径:skills/code-tour
安装命令:
npx skills add https://github.com/github/awesome-copilot --skill code-tour
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/github/awesome-copilot --skill code-tour

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

提供代码导览功能,帮助理解复杂项目的结构与逻辑流程。适用于新成员上手、代码审查与架构分析场景。使用 npx skills add 从 awesome-copilot 仓库安装,依赖项目源码可读性。访问私有仓库需确认 token 权限与代码可见范围。

code-tour 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/code-tour

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/github/awesome-copilot --skill code-tour;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

code-tour 怎么安装?

在终端执行 npx skills add https://github.com/github/awesome-copilot --skill code-tour。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

code-tour 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

code-tour 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

code-tour 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

code-tour 用的人多吗?

公开数据是累计安装 9,914 次,近一周安装 405 次,下载 3,208 次,GitHub 31,737 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

code-tour 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 来源仓库有 31,737 星,维护活跃度相对可查。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
code-tour

代码之旅 · 本页

9,91431,737需要联网未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Code Tour Skill

You are creating a CodeTour — a persona-targeted, step-by-step walkthrough of a codebase that links directly to files and line numbers. CodeTour files live in .tours/ and work with the VS Code CodeTour extension.

Two scripts are bundled in scripts/:

  • scripts/validate_tour.py — run after writing any tour. Checks JSON validity, file/directory existence, line numbers within bounds, pattern matches, nextTour cross-references, and narrative arc. Run it: python ~/.agents/skills/code-tour/scripts/validate_tour.py.tours/<name>.tour --repo-root.
  • scripts/generate_from_docs.py — when the user asks to generate from README/docs, run this first to extract a skeleton, then fill it in. Run it: python ~/.agents/skills/code-tour/scripts/generate_from_docs.py --persona new-joiner --output.tours/skeleton.tour

Two reference files are bundled:

  • references/codetour-schema.json — the authoritative JSON schema. Read it to verify any field name or type. Every field you use must conform to it.
  • references/examples.md — 8 real-world CodeTour tours from production repos with annotated techniques. Read it when you want to see how a specific feature (commands, selection, view, pattern, isPrimary, multi-tour series) is used in practice.

Real-world .tour files on GitHub

These are confirmed production .tour files. Fetch one when you need a working example of a specific step type, tour-level field, or narrative structure — don't write from memory when the real thing is one fetch away.

Find more with the GitHub code search: https://github.com/search?q=path%3A**%2F*.tour+&type=code

By step type / technique demonstrated
What to studyFile URL
directory + file+line (contributor onboarding)https://github.com/coder/code-server/blob/main/.tours/contributing.tour
selection + file+line + intro content step (accessibility project)https://github.com/a11yproject/a11yproject.com/blob/main/.tours/code-tour.tour
Minimal tutorial — tight file+line narration for interactive learninghttps://github.com/lostintangent/rock-paper-scissors/blob/master/main.tour
Multi-tour repo with nextTour chaining (cloud native OCI walkthroughs)https://github.com/lucasjellema/cloudnative-on-oci-2021/blob/main/.tours/introduction.tour
isPrimary: true (marks the onboarding entry point)https://github.com/nickvdyck/webbundlr/blob/main/.tours/getting-started.tour
pattern instead of line (regex-anchored steps)https://github.com/nickvdyck/webbundlr/blob/main/.tours/architecture.tour

Raw content tip: Prefix raw.githubusercontent.com and drop /blob/ for raw JSON access.

A great tour is not just annotated files. It is a narrative — a story told to a specific person about what matters, why it matters, and what to do next. Your goal is to write the tour that the right person would wish existed when they first opened this repo.

CRITICAL: Only create .tour JSON files. Never create, modify, or scaffold any other files.


Step 1: Discover the repo

Before asking the user anything, explore the codebase:

  • List the root directory, read the README, and check key config files (package.json, pyproject.toml, go.mod, Cargo.toml, composer.json, etc.)
  • Identify the language(s), framework(s), and what the project does
  • Map the folder structure 1–2 levels deep
  • Find entry points: main files, index files, app bootstrapping
  • Note which files actually exist — every path you write in the tour must be real

If the repo is sparse or empty, say so and work with what exists.

If the user says "generate from README" or "use the docs": run the skeleton generator first, then fill in every [TODO:...] by reading the actual files:

python skills/code-tour/scripts/generate_from_docs.py \
  --persona new-joiner \
  --output .tours/skeleton.tour

Entry points by language/framework

Don't read everything — start here, then follow imports.

StackEntry points to read first
Node.js / TSindex.js/ts, server.js, app.js, src/main.ts, package.json (scripts)
Pythonmain.py, app.py, __main__.py, manage.py (Django), app/__init__.py (Flask/FastAPI)
Gomain.go, cmd/<name>/main.go, internal/
Rustsrc/main.rs, src/lib.rs, Cargo.toml
Java / Kotlin*Application.java, src/main/java/.../Main.java, build.gradle
Rubyconfig/application.rb, config/routes.rb, app/controllers/application_controller.rb
PHPindex.php, public/index.php, bootstrap/app.php (Laravel)

Repo type variants — adjust focus accordingly

The same persona asks for different things depending on what kind of repo this is:

Repo typeWhat to emphasizeTypical anchor files
Service / APIRequest lifecycle, auth, error contractsrouter, middleware, handler, schema
Library / SDKPublic API surface, extension points, versioningindex/exports, types, changelog
CLI toolCommand parsing, config loading, output formattingmain, commands/, config
MonorepoPackage boundaries, shared contracts, build graphroot package.json/pnpm-workspace, shared/, packages/
FrameworkPlugin system, lifecycle hooks, escape hatchescore/, plugins/, lifecycle
Data pipelineSource → transform → sink, schema ownershipingest/, transform/, schema/, dbt models
Frontend appComponent hierarchy, state management, routingpages/, store/, router, api/

For monorepos: identify the 2–3 packages most relevant to the persona's goal. Don't try to tour everything — open the tour with a step that explains how to navigate the workspace, then stay focused.

Large repo strategy

For repos with 100+ files: don't try to read everything.

  1. Read entry points and the README first
  2. Build a mental model of the top 5–7 modules
  3. For the requested persona, identify the 2–3 modules that matter most and read those deeply
  4. For modules you're not covering, mention them in the intro step as "out of scope for this tour"
  5. Use directory steps for areas you mapped but didn't read — they orient without requiring full knowledge

A focused 10-step tour of the right files beats a scattered 25-step tour of everything.


Step 2: Read the intent — infer everything you can, ask only what you can't

One message from the user should be enough. Read their request and infer persona, depth, and focus before asking anything.

Intent map

User says→ Persona→ Depth→ Action
"tour for this PR" / "PR review" / "#123"pr-reviewerstandardAdd uri step for the PR; use ref for the branch
"why did X break" / "RCA" / "incident"rca-investigatorstandardTrace the failure causality chain
"debug X" / "bug tour" / "find the bug"bug-fixerstandardEntry → fault points → tests
"onboarding" / "new joiner" / "ramp up"new-joinerstandardDirectories, setup, business context
"quick tour" / "vibe check" / "just the gist"vibecoderquick5–8 steps, fast path only
"explain how X works" / "feature tour"feature-explainerstandardUI → API → backend → storage
"architecture" / "tech lead" / "system design"architectdeepBoundaries, decisions, tradeoffs
"security" / "auth review" / "trust boundaries"security-reviewerstandardAuth flow, validation, sensitive sinks
"refactor" / "safe to extract?"refactorerstandardSeams, hidden deps, extraction order
"performance" / "bottlenecks" / "slow path"performance-optimizerstandardHot path, N+1, I/O, caches
"contributor" / "open source onboarding"external-contributorquickSafe areas, conventions, landmines
"concept" / "explain pattern X"concept-learnerstandardConcept → implementation → rationale
"test coverage" / "where to add tests"test-writerstandardContracts, seams, coverage gaps
"how do I call the API"api-consumerstandardPublic surface, auth, error semantics

Infer silently: persona, depth, focus area, whether to add uri/ref, isPrimary.

Ask only if you genuinely can't infer:

  • "bug tour" but no bug described → ask for the bug description
  • "feature tour" but no feature named → ask which feature
  • "specific files" explicitly requested → honor them as required stops

Never ask about nextTour, commands, when, or stepMarker unless the user mentioned them.

PR tour recipe

For PR tours: set "ref" to the branch, open with a uri step for the PR, cover changed files first, then unchanged-but-critical files, close with a reviewer checklist.

User-provided customization — always honor these

User saysWhat to do
"cover src/auth.ts and config/db.yml"Those files are required stops
"pin to the v2.3.0 tag" / "this commit: abc123"Set "ref": "v2.3.0"
"link to PR #456" / pastes a URLAdd a uri step at the right narrative moment
"lead into the security tour when done"Set "nextTour": "Security Review"
"make this the main onboarding tour"Set "isPrimary": true
"open a terminal at this step"Add "commands": ["workbench.action.terminal.focus"]
"deep" / "thorough" / "5 steps" / "quick"Override depth accordingly

Step 3: Read the actual files — no exceptions

Every file path and line number in the tour must be verified by reading the file. A tour pointing to the wrong file or a non-existent line is worse than no tour.

For every planned step:

  1. Read the file
  2. Find the exact line of the code you want to highlight
  3. Understand it well enough to explain it to the target persona

If a user-requested file doesn't exist, say so — don't silently substitute another.


Step 4: Write the tour

Save to .tours/<persona>-<focus>.tour. Read references/codetour-schema.json for the authoritative field list. Every field you use must appear in that schema.

Tour root

{
  "$schema": "https://aka.ms/codetour-schema",
  "title": "Descriptive Title — Persona / Goal",
  "description": "One sentence: who this is for and what they'll understand after.",
  "ref": "main",
  "isPrimary": false,
  "nextTour": "Title of follow-up tour",
  "steps": []
}

Omit any field that doesn't apply to this tour.

when — conditional display. A JavaScript expression evaluated at runtime. Only show this tour if the condition is true. Useful for persona-specific auto-launching, or hiding advanced tours until a simpler one is complete.

{ "when": "workspaceFolders[0].name === 'api'" }

stepMarker — embed step anchors directly in source code comments. When set, CodeTour looks for // <stepMarker> comments in files and uses them as step positions instead of (or alongside) line numbers. Useful for tours on actively changing code where line numbers shift constantly. Example: set "stepMarker": "CT" and put // CT in the source file. Don't suggest this unless the user asks — it requires editing source files, which is unusual.


Step types — full reference

All step types: content (intro/closing, max 2), directory, file+line (workhorse), selection (code block), pattern (regex match), uri (external link), view (focus VS Code panel), commands (run VS Code commands).

Path rule: "file" and "directory" must be relative to repo root. No absolute paths, no leading ./.

When to use each step type

SituationStep type
Tour intro or closingcontent
"Here's what lives in this folder"directory
One line tells the whole storyfile + line
A function/class body is the pointselection
Line numbers shift, file is volatilepattern
PR / issue / doc gives the "why"uri
Reader should open terminal or explorerview or commands

Step count calibration

Match steps to depth and persona. These are targets, not hard limits.

DepthTotal stepsCore path stepsNotes
Quick5–83–5Vibecoder, fast explorer — cut ruthlessly
Standard9–136–9Most personas — breadth + enough detail
Deep14–1810–13Architect, RCA — every tradeoff surfaced

Scale with repo size too. A 3-file CLI doesn't get 15 steps. A 200-file monolith shouldn't be squeezed into 5.

Repo sizeRecommended standard depth
Tiny (< 20 files)5–8 steps
Small (20–80 files)8–11 steps
Medium (80–300 files)10–13 steps
Large (300+ files)12–15 steps (scoped to relevant subsystem)

Writing excellent descriptions — the SMIG formula

Every description should answer four questions in order. You don't need four paragraphs — but every description needs all four elements, even briefly.

S — Situation: What is the reader looking at? One sentence grounding them in context. M — Mechanism: How does this code work? What pattern, rule, or design is in play? I — Implication: Why does this matter for *this persona's goal specifically*? G — Gotcha: What would a smart person get wrong here? What's non-obvious, fragile, or surprising?

Descriptions should tell the reader something they couldn't learn by reading the file themselves. Name the pattern, explain the design decision, flag failure modes, and cross-reference related context.


Narrative arc — every tour, every persona

  1. Orientation — must be a file or directory step, never content-only. Use "file": "README.md", "line": 1 or "directory": "src" and put your welcome text in the description. A content-only first step (no file, directory, or uri) renders as a blank page in VS Code CodeTour — this is a known VS Code extension behaviour, not configurable.
  2. High-level map (1–3 directory or uri steps) — major modules and how they relate. Not every folder — just what this persona needs to know.
  3. Core path (file/line, selection, pattern, uri steps) — the specific code that matters. This is the heart of the tour. Read and narrate. Don't skim.
  4. Closing (content) — what the reader now understands, what they can do next, 2–3 suggested follow-up tours. If nextTour is set, reference it by name here.

Closing steps

Don't summarize — the reader just read it. Instead, tell them what they can now *do*, what to avoid, and suggest 2-3 follow-up tours.


The 20 personas

PersonaGoalMust coverAvoid
VibecoderGet the vibe fastEntry point, request flow, main modules. Max 8 steps.Deep dives, edge cases
New joinerStructured ramp-upDirectories, setup, business context, service boundaries.Advanced internals
Bug fixerRoot cause fastUser action → trigger → fault points. Repro hints + test locations.Architecture tours
RCA investigatorWhy did it failCausality chain, side effects, race conditions, observability.Happy path
Feature explainerOne feature end-to-endUI → API → backend → storage. Feature flags, edge cases.Unrelated features
PR reviewerReview the change correctlyChange story, invariants, risky areas, reviewer checklist. URI step for PR.Unrelated context
Security reviewerTrust boundariesAuth flow, input validation, secret handling, sensitive sinks.Unrelated business logic
RefactorerSafe restructuringSeams, hidden deps, coupling hotspots, safe extraction order.Feature explanations
External contributorContribute without breakingSafe areas, code style, architecture landmines.Deep internals
Tech lead / architectShape and rationaleModule boundaries, design tradeoffs, risk hotspots.Line-by-line walkthroughs

Designing a tour series

When a codebase is complex enough that one tour can't cover it well, design a series. The nextTour field chains them: when the reader finishes one tour, VS Code offers to launch the next automatically.

Plan the series before writing any tour. A good series has:

  • A clear escalation path (broad → narrow, orientation → deep-dive)
  • No duplicate steps between tours
  • Each tour standalone enough to be useful on its own

Set nextTour in each tour to the title of the next one (must match exactly). Each tour should be standalone enough to be useful on its own.


What CodeTour cannot do

If asked for any of these, say clearly that it's not supported — do not suggest a workaround that doesn't exist:

RequestReality
Auto-advance to next step after X secondsNot supported. Navigation is always manual — the reader clicks Next. There is no timer, delay, or autoplay step mechanic in CodeTour.
Embed a video or GIF in a stepNot supported. Descriptions are Markdown text only.
Run arbitrary shell commandsNot supported. commands only executes VS Code commands (e.g. workbench.action.terminal.focus), not shell commands.
Branch / conditional next stepNot supported. Tours are linear. when controls whether a tour is shown, not which step follows which.
Show a step without opening a filePartially — content-only steps work, but step 1 must have a file or directory anchor or VS Code shows a blank page.

Anti-patterns

Anti-patternFix
File listing — visiting files with "this file contains..."Tell a story; each step should depend on the previous one
Generic descriptionsName the specific pattern/gotcha unique to *this* codebase
Line number guessingNever write a line number you didn't verify by reading the file
Ignoring the personaCut every step that doesn't serve their specific goal
Hallucinated filesIf a file doesn't exist, skip the step

Quality checklist — verify before writing the file

  • Every file path is relative to the repo root (no leading / or ./)
  • Every file path read and confirmed to exist
  • Every line number verified by reading the file (not guessed)
  • Every directory is relative to the repo root and confirmed to exist
  • Every pattern regex would match a real line in the file
  • Every uri is a complete, real URL (https://...)
  • ref is a real branch/tag/commit if set
  • nextTour exactly matches the title of another .tour file if set
  • Only .tour JSON files created — no source code touched
  • First step has a file or directory anchor (content-only first step = blank page in VS Code)
  • Tour ends with a closing content step that tells the reader what they can *do* next
  • Every description answers SMIG — Situation, Mechanism, Implication, Gotcha
  • Persona's priorities drive step selection (cut everything that doesn't serve their goal)
  • Step count matches requested depth and repo size (see calibration table)
  • At most 2 content-only steps (intro + closing)
  • All fields conform to references/codetour-schema.json

Step 5: Validate the tour

Always run the validator immediately after writing the tour file. Do not skip this step.

python ~/.agents/skills/code-tour/scripts/validate_tour.py .tours/<name>.tour --repo-root .

The validator checks:

  • JSON validity
  • Every file path exists and every line is within file bounds
  • Every directory exists
  • Every pattern regex compiles and matches at least one line in the file
  • Every uri starts with https://
  • nextTour matches an existing tour title in .tours/
  • Content-only step count (warns if > 2)
  • Narrative arc (warns if no orientation or closing step)

Fix every error before proceeding. Re-run until the validator reports ✓ or only warnings. Warnings are advisory — use your judgment. Do not show the user the tour until validation passes.

Common VS Code issues: Content-only first step renders blank (anchor to file/directory instead). Absolute or ./-prefixed paths silently fail. Out-of-bounds line numbers scroll nowhere.

If you can't run scripts, manually verify: step 1 has file/directory, all paths exist, all line numbers are in bounds, nextTour matches exactly.

Autoplay: isPrimary: true + .vscode/settings.json with {"codetour.promptForPrimaryTour": true} prompts on repo open. Omit ref for tours that should appear on any branch.

Share: For public repos, users can open tours at https://vscode.dev/github.com/<owner>/<repo> with no install.


Step 6: Summarize

After writing the tour, tell the user:

  • File path (.tours/<name>.tour)
  • One-paragraph summary of what the tour covers and who it's for
  • The vscode.dev URL if the repo is public (so they can share it immediately)
  • 2–3 suggested follow-up tours (or the next tour in the series if one was planned)
  • Any user-requested files that didn't exist (be explicit — don't quietly substitute)

File naming

<persona>-<focus>.tour — kebab-case, communicates both:

onboarding-new-joiner.tour
bug-fixer-payment-flow.tour
architect-overview.tour
vibecoder-quickstart.tour
pr-review-auth-refactor.tour
security-auth-boundaries.tour
concept-dependency-injection.tour
rca-login-outage.tour

平台分布

Codex

37.69%
按下载量换算1,209

Claude

29.95%
按下载量换算961

Cursor

19.66%
按下载量换算631

Gemini CLI

9.92%
按下载量换算318

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills