Code Tour

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/.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.

When to Use

• When tasked with creating an onboarding guide for new repository contributors.
• To generate a step-by-step walkthrough of a complex bug fix or architectural feature.
• Whenever a user asks to "create a tour" or "make a code tour" for a specific codebase subsystem.

When Not to Use

• When writing standard README.md documentation that doesn't require interactive IDE navigation.
• If the requested files are highly volatile and change line numbers constantly (unless using pattern anchors).
• For generating automated API reference documentation.

Common Pitfalls

• Hallucinated Files: Creating a .tour file that points to non-existent files or invalid line numbers, instantly breaking the VS Code extension.
• Content-Only Starts: Opening the tour with a content step instead of a file or directory anchor, which causes VS Code to render a completely blank page.
• Absolute Paths: Writing file paths with a leading / or ./ instead of making them strictly relative to the repository root.

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:

By step type / technique demonstrated

| What to study | File URL | | --- | --- | | directory + file+line (contributor onboarding) | | | selection + file+line + intro content step (accessibility project) | | | Minimal tutorial — tight file+line narration for interactive learning | | | Multi-tour repo with nextTour chaining (cloud native OCI walkthroughs) | | | isPrimary: true (marks the onboarding entry point) | | | pattern instead of line (regex-anchored steps) | |

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 ~/.agents/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.

| Stack | Entry points to read first | | :--- | :--- | | Node.js / TS | index.js/ts, server.js, app.js, src/main.ts, package.json (scripts) | | Python | main.py, app.py, __main__.py, manage.py (Django), app/__init__.py (Flask/FastAPI) | | Go | main.go, cmd//main.go, internal/ | | Rust | src/main.rs, src/lib.rs, Cargo.toml | | Java / Kotlin | *Application.java, src/main/java/.../Main.java, build.gradle | | Ruby | config/application.rb, config/routes.rb, app/controllers/application_controller.rb | | PHP | index.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 type | What to emphasize | Typical anchor files | | :--- | :--- | :--- | | Service / API | Request lifecycle, auth, error contracts | router, middleware, handler, schema | | Library / SDK | Public API surface, extension points, versioning | index/exports, types, changelog | | CLI tool | Command parsing, config loading, output formatting | main, commands/, config | | Monorepo | Package boundaries, shared contracts, build graph | root package.json/pnpm-workspace, shared/, packages/ | | Framework | Plugin system, lifecycle hooks, escape hatches | core/, plugins/, lifecycle | | Data pipeline | Source → transform → sink, schema ownership | ingest/, transform/, schema/, dbt models | | Frontend app | Component hierarchy, state management, routing | pages/, 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-reviewer | standard | Add uri step for the PR; use ref for the branch | | "why did X break" / "RCA" / "incident" | rca-investigator | standard | Trace the failure causality chain | | "debug X" / "bug tour" / "find the bug" | bug-fixer | standard | Entry → fault points → tests | | "onboarding" / "new joiner" / "ramp up" | new-joiner | standard | Directories, setup, business context | | "quick tour" / "vibe check" / "just the gist" | vibecoder | quick | 5–8 steps, fast path only | | "explain how X works" / "feature tour" | feature-explainer | standard | UI → API → backend → storage | | "architecture" / "tech lead" / "system design" | architect | deep | Boundaries, decisions, tradeoffs | | "security" / "auth review" / "trust boundaries" | security-reviewer | standard | Auth flow, validation, sensitive sinks | | "refactor" / "safe to extract?" | refactorer | standard | Seams, hidden deps, extraction order | | "performance" / "bottlenecks" / "slow path" | performance-optimizer | standard | Hot path, N+1, I/O, caches | | "contributor" / "open source onboarding" | external-contributor | quick | Safe areas, conventions, landmines | | "concept" / "explain pattern X" | concept-learner | standard | Concept → implementation → rationale | | "test coverage" / "where to add tests" | test-writer | standard | Contracts, seams, coverage gaps | | "how do I call the API" | api-consumer | standard | Public 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 says | What 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 URL | Add 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/-.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 // 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

| Situation | Step type | | :--- | :--- | | Tour intro or closing | content | | "Here's what lives in this folder" | directory | | One line tells the whole story | file + line | | A function/class body is the point | selection | | Line numbers shift, file is volatile | pattern | | PR / issue / doc gives the "why" | uri | | Reader should open terminal or explorer | view or commands |

---

Step count calibration

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

| Depth | Total steps | Core path steps | Notes | | :--- | :--- | :--- | :--- | | Quick | 5–8 | 3–5 | Vibecoder, fast explorer — cut ruthlessly | | Standard | 9–13 | 6–9 | Most personas — breadth + enough detail | | Deep | 14–18 | 10–13 | Architect, 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 size | Recommended standard depth | | :--- | :--- | | Tiny (.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.

## References

-

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [Cogni-AI-OU](https://github.com/Cogni-AI-OU)
- **Source:** [Cogni-AI-OU/cogni-ai-agent-skills](https://github.com/Cogni-AI-OU/cogni-ai-agent-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.