spec - script

Ideas 12

Values 12

Value	Distribution	Ideas	Share
SCR: Activity bar reqlan icon — empty state CTA "Create base", or Command Palette: `Reqlan: Create Base`. Explorer: `reqs/product.rq` present, no `.reqlan` yet. VO: Installing the extension puts reqlan in the editor. It does not create a base. A base is a folder that owns a `.reqlan` directory — the boundary of one ideas graph. SCR: Sticky: "I installed the extension — why is the activity bar empty?" Point at missing `.reqlan` in Explorer. VO: Extension install and base create are two layers. Until the folder has a marker, there is no index boundary for neighbourhood views to bind to. SCR: Accept create-base at the workspace-folder root (`reqlan-tutorials-demo`). Explorer refreshes — expand `.reqlan`: `config.json`, `.rqignore`. Do not open the sqlite file. VO: That marker is shared application memory. Config and ignore rules live here. The ideas index appears under `.reqlan` once indexing activates — treat it as tooling memory, not a document you edit. SCR: Glance at activity bar or status: idea counts / readiness updating. Return to `product.rq` — caret on `welcome_flow`. VO: Same ideas you already wrote — now the tooling knows which graph they belong to. SCR: Optional ten seconds: terminal in a scratch empty folder — `npx reqlan init` — show identical `.reqlan` tree, then close terminal and return to demo folder. VO: Editor create-base and CLI init seed the same marker. Pick the host that fits the moment. SCR: Whiteboard beat — dashed box labelled "nested base" inside a parent folder. Do not create one live. VO: One base is enough for day one. Nested bases exist for larger trees — we name them later in Concepts. VO: Try this: confirm `.reqlan` is visible in Explorer, open `product.rq`, leave the caret on an idea, and watch the activity bar populate. SCR: End card → next: "Neighbourhood in the activity bar".		1	8.3%
SCR: Click the reqlan activity bar icon. Place caret inside `welcome_flow` in `reqs/product.rq`. Confirm active base label shows `reqlan-tutorials-demo` if a picker is visible. VO: The activity bar is your neighbourhood view — what sits near the idea under the cursor inside the active base — not a dump of the whole workspace. That keeps things readable when you copy a slice into chat later. SCR: Toggle "Sync with editor" on if off. Move caret from `welcome_flow` to `onboarding_checklist` — panes update. Briefly show hop depth control: 1 hop → 2 hops — neighbour list widens then tighten back to 1. VO: Sync follows your caret. Hop depth widens or tightens how far references travel. Keep it small while you are learning. Context binds to the active base's index — not a merged global graph. SCR: Sticky: "Should I open Ideas Summary every time?" Shake head — stay on activity bar. VO: No. The activity bar is for local context while you edit. Ideas Summary is for when you choose a broader map — not a habit on every keystroke. SCR: Activity bar header → Copy as Markdown (or equivalent action). Paste into an untitled scratch file — show compact markdown, not the whole `reqs/` tree. VO: Copy a focused slice when you want to talk to a model. Copying the neighbourhood keeps the answer sharp — pasting the whole requirements folder just buries the question. SCR: Command Palette → `Reqlan: Ideas Summary`. Graph or overview tab — centre on current idea, pan once, close tab or switch back to Explorer. Total on screen: under fifteen seconds. VO: Ideas Summary is for broader navigation when you choose it. Day one: glance and leave unless you are lost. VO: Try this: turn sync off, pin `welcome_flow`, click around the editor to other ideas, then turn sync back on and watch focus return. SCR: End card → next: "Ask @reqlan".		1	8.3%
SCR: Cold open on `product.rq`. Hover an unresolved `[login_api]` — Problems panel or squiggle if visible. Explorer shows both `product.rq` and `auth.rq`. VO: One file is a great start. Real projects become a web of files — and links that worked by accident in a single document need imports to stay honest. SCR: Open `auth.rq`. Show `login_api`, `session_policy`, and the `auth_core` ideaset paren list. VO: Each file is its own ideaspace. Moving auth ideas here does not change what an idea is — it changes where the handle lives. SCR: Return to `product.rq`. Type `from "./auth.rq" import login_api`. Save. Link resolves. Go-to-definition lands in `auth.rq`. VO: From-import pulls one idea into scope. Use it when the call site names that neighbour often and nothing else from the file. SCR: Add `import "./auth.rq" as auth`. Replace session reference with `[auth.session_policy]`. Brief pause on readability. VO: Namespace import gives you a prefix for the whole file. Judgement call: if you reference three ideas from auth, namespace usually reads cleaner than three from-import lines. SCR: Highlight `auth_core` ideaset in `auth.rq`. Optional: Ideas Summary ideas table filtered to `auth.rq` path. VO: Ideasets group names without inventing a heavier module system. They document membership — useful in reviews — but imports still control what this file can link to. SCR: Misconception beat — split screen: over-imported file with six from-imports vs one namespace line. Narrator picks namespace. VO: Common mistake: import everything "just in case". Import the smallest surface that keeps references readable. SCR: Move `password_rules` live from `product.rq` into `auth.rq` if still in product. Fix product with `[auth.password_rules]`. VO: Try this: move `password_rules` into `auth.rq` and fix `product.rq` with the smallest import that works. If you maintain multiple bases later, imports do not cross base boundaries — same rule as get-started. SCR: End card → "Attributes without bloat".		1	8.3%
SCR: Dark title card: reqlan wordmark + "Get started · 01 · Why requirements as code". Brand palette. VO: If you work with language models on real projects, you already write requirements — you just write them into the chat window, over and over. SCR: Split screen. Left: a bloated chat paste — one long paragraph mixing login, sessions, and audit rules. Right: three short named lines from the example. VO: reqlan is a small language for turning that intent into a graph of named ideas — so you, your teammates, and your tools can find, link, and reuse them instead of retyping. SCR: Brief montage (no narration overlap): VS Code with a `.rq` file, reqlan activity bar icon, Ideas Summary graph with a handful of nodes. VO: The day-one job is simple: name the idea, write the prose, point at related work. You can see it here — a file of names, and lines that connect them. SCR: Show a sticky note or chat bubble: "Isn't this just another notes app?" Cut to the named-ideas side of the contrast — three handles you can reference by name. VO: Notes apps store paragraphs. reqlan stores handles. You do not need a schema on day one — you need names your tools can search and link. SCR: Browser: Visual Studio Marketplace page for the reqlan extension — or Open VSX equivalent. Click Install. Switch to VS Code: Extensions view confirms install. File → Open Folder → create or select `reqlan-tutorials-demo`. VO: Install the reqlan extension from the Marketplace or Open VSX. Open a folder — even an empty one — so the editor has a workspace root. SCR: Onboarding webview appears: "Thanks for installing". Pause on docs / marketplace links in the welcome body. Do not create a base yet. VO: On first launch you get a short welcome screen. Keep those links handy. Creating a base — the `.reqlan` marker — comes after you have written a few ideas. SCR: End card: CTA row — Install · Docs · next episode "Your first idea". VO: Next: write your first idea in under a minute.		1	8.3%
SCR: Explorer in `reqlan-tutorials-demo`. Right-click → New Folder `reqs`. New File `reqs/product.rq`. Caret at line 1. VO: A reqlan file is just text with a `.rq` suffix. Start with the smallest form: a name, then the thought. SCR: Type keystroke-by-keystroke: `product_name reqlan tutorials demo`. Pause so token colours settle — name vs body. VO: That one line is already an idea. The name is a handle you will reference later. When you need more room, open a block. SCR: New line. Type the `welcome_flow` block from the example — open brace, body prose, close brace. Soft zoom on the body text inside braces. VO: Inside the braces, the first free text is the main description. No required metadata yet — just a name and the thought. SCR: Status bar: language mode shows reqlan (or Reqlan). Problems panel empty. Brief hover on bracket/brace highlighting if visible. VO: Save the file. If the extension is active, language support arrives immediately — highlighting, not a separate compile step. SCR: Chat bubble or sticky: "Do I need YAML front matter or a schema?" Cut back to the two-line file — no `@` attributes, no imports. VO: You do not. One-liners and blocks are valid on their own. Attributes and imports come later when you feel the need — not before your first idea exists. VO: Try this: rename `welcome_flow` to something that matches your own product, and rewrite the sentence in your voice. Save again. SCR: End card → next: "Link ideas".		1	8.3%
SCR: New file `src/login.ts` — paste the stub from the example. New file `reqs/auth.rq` — type `login_api` block with file reference `Implementation: ["./src/login.ts"]` and link to `[password_rules]`. VO: Requirements earn their keep when they touch reality. Point an idea at a file with a file reference — the path lives in brackets like any other link. SCR: Ctrl/Cmd-click `["./src/login.ts"]` inside `login_api` — land in `login.ts`. Brief pause on the opened tab. VO: Same navigation habit as idea links — the graph knows about workspace files, not only `.rq` sources. SCR: Sticky: "File refs only go one way?" Place caret inside `login.ts` on the empty-password guard. Open activity bar — show requirements referencing this file or neighbourhood reacting to non-`.rq` focus. VO: Context scope works on any workspace file, not only `.rq` sources. Stand in the implementation and related requirements should still surface — the reverse path. SCR: Optional five seconds: mention folder references like `["./media/interviews/"]` on a slide — do not create live. VO: Folder references work the same way when assets span a directory. SCR: Recap montage (five seconds): seven episode titles stacked — Why · First idea · Link · Base · Activity bar · Ask · Point at code. VO: You now have the get-started loop: name ideas, link them, create a base, inspect the neighbourhood, ask the tools with a small slice, and stitch requirements to code. VO: Try this before you go: link one more source file your welcome flow actually depends on — a component, route, or test. SCR: End card: "Concepts · The building blocks, the extension map, then the CLI" — series handoff to [concepts_series].		1	8.3%
SCR: Open `reqs/product.rq`. `welcome_flow` and `product_name` visible. Caret in the welcome_flow body. VO: Ideas earn power when they point at each other. Square brackets name a neighbour in the graph. SCR: Below the welcome prose, add a new block `onboarding_checklist` from the example. Inside `welcome_flow`, type `Success continues into [onboarding_checklist].` VO: The bracket link creates an edge. The target idea can live in the same file or another — the syntax is the same. SCR: F12 or Ctrl/Cmd-click `[onboarding_checklist]` — jump to the new block. Then Peek Definition (Alt+F12) back to the call site. VO: Follow the link — you are navigating a graph, not scrolling a long document. Peek when you want context without losing your place. SCR: In `onboarding_checklist`, add the wikilink: `Prefer [[welcome_flow\|the welcome screen]] as the way back if they bail early.` VO: Double-bracket wikilinks let you show a readable label right in the prose. The alias is for reading; the target is still the idea name `welcome_flow`. SCR: Sticky note: "Are wikilinks a different graph?" Show both link forms resolving to the same idea in go-to-definition. VO: Same graph, two spellings. Brackets for a quick hop; wikilinks when you want a readable label in prose. SCR: Optional five seconds: Command Palette → Reqlan: Show Local Graph if discoverable; otherwise stay in editor. VO: Try this: add a third idea `account_model` — accounts need an email and display name — and link both `welcome_flow` and `onboarding_checklist` to it once each. SCR: End card → next: "Your first base".		1	8.3%
SCR: Open `reqs/product.rq`. If `password_rules` is absent, add the block from the example and save. Open VS Code Chat (or Cursor Chat if recording host). VO: reqlan meets AI through focused tools — search, scoped context, small builds — not by shoving the whole graph into the prompt. SCR: VS Code Chat: type `@reqlan /rq-search password empty`. Show compact results — idea name, file path, one-line summary. Cursor alternate: invoke skill `/rq-search password empty` from chat. VO: Slash commands and skills use the `rq-` prefix. Search returns a short list on purpose — enough to pick, not enough to drown the thread. SCR: Sticky: "Why not paste product.rq?" Show result list vs a full-file paste thumbnail — reject the paste. VO: Search finds the handle. Add-to-context ships the slice. Meaning search beats scrolling when the word lives in body prose, not the idea name. SCR: Run add-to-context on `password_rules` — chat participant menu or `@reqlan` add action. Show only that idea's body joined to the thread, not the whole file. VO: Add what you need. Leave the rest in the index. SCR: Optional fifteen seconds: `@reqlan /rq-build-requirement` or Cursor skill `rq-build-requirement` with intent "rate-limit failed logins" — accept draft appended to `product.rq` as a new named idea; edit one line on camera. VO: Building a requirement from intent should land as a named idea you keep owning in source control — the model proposes; the `.rq` file remains the source of truth. VO: Try this: search for a word that only appears in a body sentence — like "twelve" inside `password_rules` — not in any idea name. SCR: End card → next: "Point ideas at code".		1	8.3%
SCR: Title card — Concepts · 01 · The building blocks. Brand palette. Quiet list of seven names fades in one at a time. VO: You already wrote ideas, linked them, created a base, inspected the neighbourhood, and asked the tools. Now let's put names to the things you used, so you and your teammates talk about them the same way. One name per beat. SCR: Open `reqs/product.rq`. Highlight `welcome_flow` one-liner and block side by side. VO: First: the idea. An idea is a named handle plus prose — one thing you care about. A one-liner is already an idea. A block gives the body more room. You do not type `@body` — the first unmarked prose inside a block simply is the body. SCR: Open `reqs/auth.rq`. Soft zoom on the auth_core ideaset paren list. VO: Next: the ideaset. An ideaset groups idea names inside parentheses. It lists membership — a light way to say "these belong together", not a second syntax to learn. SCR: Explorer: click `auth.rq` in the tree. Brief file-tab chrome — path visible, ideas inside. VO: Next: the file. A `.rq` file holds ideas and is itself a group of them — the file is part of the graph, not just a spot on disk. When imports arrive in Advanced, files stay clear about where each name lives. SCR: Explorer: expand `.reqlan/`. Whiteboard overlay or dashed box: `project/.reqlan/` vs `packages/lib/.reqlan/` — do not live-switch repos on camera. VO: Next: the base. A base is the boundary of one graph of ideas. A folder is a base only when it owns a `.reqlan` folder. Nested bases are separate graphs; the parent index stops at the child boundary. Day-to-day you work in one active base — nested trees are for larger projects, not day-two homework. SCR: Split editor: bracket link `[session_policy]` beside file reference `["./src/login.ts"]`. Optional three-second flash of `// rq:[password_rules]` in `login.ts` — no deep dive. VO: Next: the reference. A reference links an idea to another idea, a file, a folder, or a symbol. Square brackets are the same habit you already used in get-started. A non-`.rq` file you point at still shows up in the index — a neighbour, not an orphan. SCR: Cursor on `@status` and `@tags` inside `login_api`. Do not open settings panels. VO: Next: the attribute. Attributes are optional metadata you opt into — status, tags, plans, your own keys. They are not the main prose. And one thing to keep straight: `.reqlan/config.json` is not the same place as your editor settings — graph memory and editor preferences live apart. SCR: Ideas Summary → Graph tab, already centred on `welcome_flow` with hop depth 1. Pan once; close. VO: Last: the map. Ideas Summary shows a focused map view of the graph — filtered, centred, depth-limited — built around the question you have right now. It is not a picture of everything at once. Advanced goes deeper on map craft. SCR: New scratch file `reqs/scrap-ontology.rq` — type seven one-line definitions live. VO: Try this: on a scrap page, write one sentence for each name — idea, ideaset, file, base, reference, attribute, and map. Use your own words; if a sentence needs a second clause, the idea is still fuzzy. SCR: End card — "Next: where these live in the extension." Logo + site link from [phonebook]. VO: Next episode shows where each of these appears across the extension.		1	8.3%
SCR: Title card — Concepts · 02 · Extension tour. Five blank tiles; first tile highlights. VO: You have the names. This episode shows where each one appears in the product. The extension is the host — language support plus views over the same graph the CLI and chat use. Five surfaces. One job each. Depth waits in Advanced. SCR: Tile 1 — editor with `product.rq`. Type one character; show highlighting. Ctrl/Cmd-click a bracket link; peek definition. VO: Surface one: language support. Open this when you are authoring or navigating `.rq` source. Ideas and references live in files — not in a sidebar form. Highlighting, go-to-definition, and bracket navigation are how the nouns stay honest in source control. SCR: Tile 2 — click reqlan activity bar. Caret on `welcome_flow`. Sync on; hop depth 1; copy-as-markdown flash. VO: Surface two: activity bar. Open this when the question is "what sits near this idea inside the active base?" Neighbourhood context, base picker when multiple bases exist, hop depth, copy slice — the local view you felt in get-started, not a workspace-wide dump. SCR: Tile 3 — Command Palette: Reqlan: Open Ideas Summary. Overview glance; Ideas tab with status column; Graph tab one pan — close without opening every tab. VO: Surface three: Ideas Summary. Open this when you need a catalog, a table, or a focused map. Its tabs mirror ideas, ideasets, references, attributes, and bases — use them when the question needs a list or a map, not as a mid-edit habit. If you are typing prose, do it in the editor; open Summary when you need to look something up. SCR: Tile 4 — split beat. Chat: `@reqlan /rq-search empty password`. Command Palette: type `Reqlan:` — show category list. Optional Cursor: `rq-search` skill name flash. VO: Surface four: chat, skills, MCP, and command palette. Open this when you want a focused graph operation — search, add-to-context, export, open Summary — the same small-slice habit as get-started. Skills use the `rq-` prefix; MCP tools share the same index. The palette is where you look when you forgot the exact slash command. SCR: Tile 5 — Command Palette: export HTML form header only, collapsed. Terminal: `npx reqlan search "empty password"` — show three lines of output; return to editor. VO: Surface five: export and CLI. Open this when the graph must leave the editor — static HTML per [html_export], headless search per [cli_package] and [commands]. Extension and terminal share one analytical core; pick the host that fits the moment. The next concepts pack walks the CLI commands; Advanced covers graph craft, MCP skills, and CI judgement. SCR: Five tiles visible. On-screen quiz prompts appear one at a time — no clicking yet. VO: Try this — surface quiz, say it aloud before you click: one — copy a neighbourhood slice for chat; two — filter every idea marked draft; three — search by meaning when you forgot the idea name. Answers: activity bar, Ideas Summary tables, chat or palette search. If you hesitated on two, that is the Ideas Summary job — not the activity bar. SCR: End card — Concepts · CLI thumbnail. CTA from [phonebook]. VO: Map first. Next we name the terminal surface — then Advanced when you are ready to craft.		1	8.3%
SCR: Title card — Concepts · 03 · CLI. Brand palette. Quiet terminal prompt at `reqlan-tutorials-demo`. VO: The extension tour named five surfaces. This pack is surface five up close — the [cli_package] binary `reqlan` (alias `rq`) on the same analytical core and `.reqlan` index. SCR: Terminal: `npx @reqlan/cli --help` — flash command list. Optional: mention global `npm i -g @reqlan/cli`. VO: Install is a package, not a Marketplace button. Use `npx` for a one-shot, or install globally when the terminal is your daily host. SCR: Explorer: `.reqlan/` already present from get-started. Brief beat — do not recreate live unless demo was reset. Optional alternate: empty sibling folder + `npx reqlan init` showing the marker appear. VO: [init] seeds the same `.reqlan` tree as Create Base in the editor. Extension install and base create stay separate layers — the CLI only owns the base marker path. SCR: `npx reqlan search "empty password"` — compact hits. Then `npx reqlan analyse --idea login_api`. VO: Search stays concise on purpose — same [llm_first] discipline as chat. Analyse expands one idea: references, files, attributes — not a whole-graph JSON dump unless you ask. SCR: `npx reqlan parse reqs/auth.rq` clean. Mention `parse reqs/ --fail-on-diagnostics` as the CI gate pattern. VO: Parse is how CI fails the build when requirements stop parsing — same sources the extension indexes. SCR: Ten-second export beat: `npx reqlan export --out ./docs-site` — same pipeline as the export form per [html_export]. VO: Export is one [commands] verb among equals. Form for exploration; CLI for scripts and pipelines. SCR: Slide: `--cwd` / `REQLAN_WORKSPACE`, `--json`, one active base per run. Explorer flash on nested-base dashed boundary — do not create one live. VO: Pin the workspace when the shell is in the wrong folder. Prefer `--json` for machines. Nested bases are separate runs — same craft as context and export in Advanced. SCR: Misconception beat — running from a parent monorepo without `--cwd`; results look empty or wrong. VO: If CLI and chat disagree, check cwd and active base first — not two different graphs. VO: Try this: from the demo root, run `npx reqlan parse reqs/ --fail-on-diagnostics` and imagine that line in CI. SCR: End card → Advanced overview. CTA from [phonebook].		1	8.3%
SCR: Title card: "Advanced · Overview · Beyond one file". Brand palette. Quiet montage of the demo workspace — Explorer tree, activity bar, Ideas Summary graph tab. VO: You already know the loop — name ideas, link neighbours, create a base, inspect locally, ask with a small slice. Advanced reqlan is not a second install tour. It is judgement about how graphs scale without becoming noise. SCR: Theme map slide builds one tile at a time: imports, attributes, graph maps, context, comment links, export, CLI/MCP, modelling. VO: Eight themes ahead. Imports and ideasets keep multi-file graphs honest. Attributes and plans stay slim unless they drive a view. Ideas Summary is a map you choose — not a mandatory god's-eye dump. Context and tokens carry forward the get-started habit: small slices beat whole-repo pastes. Comment references close the loop from code back to requirements. HTML export and CLI/MCP share the same analytical core — different hosts, same graph. Modelling patterns tie it together — including work that is not software. SCR: Quick beat on active base picker / `.reqlan` in Explorer — no create-base CTA. Nested base shown as dashed boundary on slide only. VO: Bases show up here as craft, not ceremony. One active base at a time. Know where your graph boundary is — nested bases are separate graphs, not folders you merge by accident. That discipline matters when you export, run CLI search, or hand context to an agent. SCR: Discipline slide: small slices, purposeful attributes, maps you choose. VO: Carry get-started forward. Token discipline is a habit, not a feature you toggle once. If an attribute never appears in how you navigate, delete it. SCR: End card: "Start with imports" — first deep dive. CTA: site tutorials index. VO: Pick your first deep dive. We recommend imports — split files before the graph outgrows one document. Every episode stands alone if you need a different entry point.		1	8.3%

Idea	Value	Status	Tags	Summary
adv_01_imports_ideasets	SCR: Cold open on `product.rq`. Hover an unresolved `[login_api]` — Problems panel or squiggle if visible. Explorer shows both `product.rq` and `auth.rq`. VO: One file is a great start. Real projects become a web of files — and links that worked by accident in a single document need imports to stay honest. SCR: Open `auth.rq`. Show `login_api`, `session_policy`, and the `auth_core` ideaset paren list. VO: Each file is its own ideaspace. Moving auth ideas here does not change what an idea is — it changes where the handle lives. SCR: Return to `product.rq`. Type `from "./auth.rq" import login_api`. Save. Link resolves. Go-to-definition lands in `auth.rq`. VO: From-import pulls one idea into scope. Use it when the call site names that neighbour often and nothing else from the file. SCR: Add `import "./auth.rq" as auth`. Replace session reference with `[auth.session_policy]`. Brief pause on readability. VO: Namespace import gives you a prefix for the whole file. Judgement call: if you reference three ideas from auth, namespace usually reads cleaner than three from-import lines. SCR: Highlight `auth_core` ideaset in `auth.rq`. Optional: Ideas Summary ideas table filtered to `auth.rq` path. VO: Ideasets group names without inventing a heavier module system. They document membership — useful in reviews — but imports still control what this file can link to. SCR: Misconception beat — split screen: over-imported file with six from-imports vs one namespace line. Narrator picks namespace. VO: Common mistake: import everything "just in case". Import the smallest surface that keeps references readable. SCR: Move `password_rules` live from `product.rq` into `auth.rq` if still in product. Fix product with `[auth.password_rules]`. VO: Try this: move `password_rules` into `auth.rq` and fix `product.rq` with the smallest import that works. If you maintain multiple bases later, imports do not cross base boundaries — same rule as get-started. SCR: End card → "Attributes without bloat".			Audience: builders with a single-file graph who feel reference sprawl or duplicate idea names across files. Outcome: split ideas across files; use `from` / `import` and namespace import; group with ideasets; resolve qualified references without breaking call-site readability. Runtime: 9 – 10 min. Demo: advanced baseline open on `reqs/product.rq`; `reqs/auth.rq` exists with `login_api` and `session_policy` but `product.rq` still references them without imports — caret on a broken or unresolved link. Related: [import_from] [import_namespace] [ideaset] [file] [reference]
adv_overview	SCR: Title card: "Advanced · Overview · Beyond one file". Brand palette. Quiet montage of the demo workspace — Explorer tree, activity bar, Ideas Summary graph tab. VO: You already know the loop — name ideas, link neighbours, create a base, inspect locally, ask with a small slice. Advanced reqlan is not a second install tour. It is judgement about how graphs scale without becoming noise. SCR: Theme map slide builds one tile at a time: imports, attributes, graph maps, context, comment links, export, CLI/MCP, modelling. VO: Eight themes ahead. Imports and ideasets keep multi-file graphs honest. Attributes and plans stay slim unless they drive a view. Ideas Summary is a map you choose — not a mandatory god's-eye dump. Context and tokens carry forward the get-started habit: small slices beat whole-repo pastes. Comment references close the loop from code back to requirements. HTML export and CLI/MCP share the same analytical core — different hosts, same graph. Modelling patterns tie it together — including work that is not software. SCR: Quick beat on active base picker / `.reqlan` in Explorer — no create-base CTA. Nested base shown as dashed boundary on slide only. VO: Bases show up here as craft, not ceremony. One active base at a time. Know where your graph boundary is — nested bases are separate graphs, not folders you merge by accident. That discipline matters when you export, run CLI search, or hand context to an agent. SCR: Discipline slide: small slices, purposeful attributes, maps you choose. VO: Carry get-started forward. Token discipline is a habit, not a feature you toggle once. If an attribute never appears in how you navigate, delete it. SCR: End card: "Start with imports" — first deep dive. CTA: site tutorials index. VO: Pick your first deep dive. We recommend imports — split files before the graph outgrows one document. Every episode stands alone if you need a different entry point.			Audience: viewers who finished get-started and concepts ( or equivalent ) and want the map of craft themes before committing to a deep dive. Outcome: name the eight advanced themes, understand how they connect to get-started discipline, and pick a first episode. Runtime: 3 – 4 min. Demo: `reqlan-tutorials-demo` at advanced baseline — multi-file graph, `.reqlan` present, no new files typed; montage only. Related: [advanced_series_brief] [general_purpose] [llm_first] [concepts_series] [base] [multi_base_environment]
con_01_ontology	SCR: Title card — Concepts · 01 · The building blocks. Brand palette. Quiet list of seven names fades in one at a time. VO: You already wrote ideas, linked them, created a base, inspected the neighbourhood, and asked the tools. Now let's put names to the things you used, so you and your teammates talk about them the same way. One name per beat. SCR: Open `reqs/product.rq`. Highlight `welcome_flow` one-liner and block side by side. VO: First: the idea. An idea is a named handle plus prose — one thing you care about. A one-liner is already an idea. A block gives the body more room. You do not type `@body` — the first unmarked prose inside a block simply is the body. SCR: Open `reqs/auth.rq`. Soft zoom on the auth_core ideaset paren list. VO: Next: the ideaset. An ideaset groups idea names inside parentheses. It lists membership — a light way to say "these belong together", not a second syntax to learn. SCR: Explorer: click `auth.rq` in the tree. Brief file-tab chrome — path visible, ideas inside. VO: Next: the file. A `.rq` file holds ideas and is itself a group of them — the file is part of the graph, not just a spot on disk. When imports arrive in Advanced, files stay clear about where each name lives. SCR: Explorer: expand `.reqlan/`. Whiteboard overlay or dashed box: `project/.reqlan/` vs `packages/lib/.reqlan/` — do not live-switch repos on camera. VO: Next: the base. A base is the boundary of one graph of ideas. A folder is a base only when it owns a `.reqlan` folder. Nested bases are separate graphs; the parent index stops at the child boundary. Day-to-day you work in one active base — nested trees are for larger projects, not day-two homework. SCR: Split editor: bracket link `[session_policy]` beside file reference `["./src/login.ts"]`. Optional three-second flash of `// rq:[password_rules]` in `login.ts` — no deep dive. VO: Next: the reference. A reference links an idea to another idea, a file, a folder, or a symbol. Square brackets are the same habit you already used in get-started. A non-`.rq` file you point at still shows up in the index — a neighbour, not an orphan. SCR: Cursor on `@status` and `@tags` inside `login_api`. Do not open settings panels. VO: Next: the attribute. Attributes are optional metadata you opt into — status, tags, plans, your own keys. They are not the main prose. And one thing to keep straight: `.reqlan/config.json` is not the same place as your editor settings — graph memory and editor preferences live apart. SCR: Ideas Summary → Graph tab, already centred on `welcome_flow` with hop depth 1. Pan once; close. VO: Last: the map. Ideas Summary shows a focused map view of the graph — filtered, centred, depth-limited — built around the question you have right now. It is not a picture of everything at once. Advanced goes deeper on map craft. SCR: New scratch file `reqs/scrap-ontology.rq` — type seven one-line definitions live. VO: Try this: on a scrap page, write one sentence for each name — idea, ideaset, file, base, reference, attribute, and map. Use your own words; if a sentence needs a second clause, the idea is still fuzzy. SCR: End card — "Next: where these live in the extension." Logo + site link from [phonebook]. VO: Next episode shows where each of these appears across the extension.			Audience: viewers who finished get-started ( or equivalent ) and want shared vocabulary without repeating the install loop. Outcome: define each core ontology noun in one sentence; place [base] as graph boundary ( including nesting ); distinguish [attribute_body] from optional `@` metadata; name [cartographic_map] as a chosen cut, not a mandatory dump. Runtime: 10 – 12 min. Demo: `reqlan-tutorials-demo` at get-started end state — `.reqlan/` present; `reqs/product.rq`, `reqs/auth.rq`, `src/login.ts`; caret starts on `welcome_flow` in `product.rq`. Related: [idea] [ideaset] [file] [base] [base_detail] [reference] [attribute] [attribute_body] [cartographic_map] [referenced_files] [configuration] [base_nesting] [multi_base_environment]
con_02_extension_tour	SCR: Title card — Concepts · 02 · Extension tour. Five blank tiles; first tile highlights. VO: You have the names. This episode shows where each one appears in the product. The extension is the host — language support plus views over the same graph the CLI and chat use. Five surfaces. One job each. Depth waits in Advanced. SCR: Tile 1 — editor with `product.rq`. Type one character; show highlighting. Ctrl/Cmd-click a bracket link; peek definition. VO: Surface one: language support. Open this when you are authoring or navigating `.rq` source. Ideas and references live in files — not in a sidebar form. Highlighting, go-to-definition, and bracket navigation are how the nouns stay honest in source control. SCR: Tile 2 — click reqlan activity bar. Caret on `welcome_flow`. Sync on; hop depth 1; copy-as-markdown flash. VO: Surface two: activity bar. Open this when the question is "what sits near this idea inside the active base?" Neighbourhood context, base picker when multiple bases exist, hop depth, copy slice — the local view you felt in get-started, not a workspace-wide dump. SCR: Tile 3 — Command Palette: Reqlan: Open Ideas Summary. Overview glance; Ideas tab with status column; Graph tab one pan — close without opening every tab. VO: Surface three: Ideas Summary. Open this when you need a catalog, a table, or a focused map. Its tabs mirror ideas, ideasets, references, attributes, and bases — use them when the question needs a list or a map, not as a mid-edit habit. If you are typing prose, do it in the editor; open Summary when you need to look something up. SCR: Tile 4 — split beat. Chat: `@reqlan /rq-search empty password`. Command Palette: type `Reqlan:` — show category list. Optional Cursor: `rq-search` skill name flash. VO: Surface four: chat, skills, MCP, and command palette. Open this when you want a focused graph operation — search, add-to-context, export, open Summary — the same small-slice habit as get-started. Skills use the `rq-` prefix; MCP tools share the same index. The palette is where you look when you forgot the exact slash command. SCR: Tile 5 — Command Palette: export HTML form header only, collapsed. Terminal: `npx reqlan search "empty password"` — show three lines of output; return to editor. VO: Surface five: export and CLI. Open this when the graph must leave the editor — static HTML per [html_export], headless search per [cli_package] and [commands]. Extension and terminal share one analytical core; pick the host that fits the moment. The next concepts pack walks the CLI commands; Advanced covers graph craft, MCP skills, and CI judgement. SCR: Five tiles visible. On-screen quiz prompts appear one at a time — no clicking yet. VO: Try this — surface quiz, say it aloud before you click: one — copy a neighbourhood slice for chat; two — filter every idea marked draft; three — search by meaning when you forgot the idea name. Answers: activity bar, Ideas Summary tables, chat or palette search. If you hesitated on two, that is the Ideas Summary job — not the activity bar. SCR: End card — Concepts · CLI thumbnail. CTA from [phonebook]. VO: Map first. Next we name the terminal surface — then Advanced when you are ready to craft.			Audience: viewers who know the ontology nouns and want a calm map of where they appear in the product. Outcome: name the primary [extension] surfaces and which job each owns; know where not to dig on day two. Runtime: 8 – 10 min. Demo: full get-started end state of `reqlan-tutorials-demo` ( `.reqlan`, `reqs/product.rq`, `reqs/auth.rq`, `src/login.ts` ); activity bar closed; Ideas Summary closed; chat available. Related: [extension] [activity_bar] [ideas_summary] [command_palette] [ai_integration] [html_export] [graph_view] [local_graph_view] [simple_views] [chat_skill_naming] [installation] [mcp_tools] [cli_package] [commands]
con_03_cli	SCR: Title card — Concepts · 03 · CLI. Brand palette. Quiet terminal prompt at `reqlan-tutorials-demo`. VO: The extension tour named five surfaces. This pack is surface five up close — the [cli_package] binary `reqlan` (alias `rq`) on the same analytical core and `.reqlan` index. SCR: Terminal: `npx @reqlan/cli --help` — flash command list. Optional: mention global `npm i -g @reqlan/cli`. VO: Install is a package, not a Marketplace button. Use `npx` for a one-shot, or install globally when the terminal is your daily host. SCR: Explorer: `.reqlan/` already present from get-started. Brief beat — do not recreate live unless demo was reset. Optional alternate: empty sibling folder + `npx reqlan init` showing the marker appear. VO: [init] seeds the same `.reqlan` tree as Create Base in the editor. Extension install and base create stay separate layers — the CLI only owns the base marker path. SCR: `npx reqlan search "empty password"` — compact hits. Then `npx reqlan analyse --idea login_api`. VO: Search stays concise on purpose — same [llm_first] discipline as chat. Analyse expands one idea: references, files, attributes — not a whole-graph JSON dump unless you ask. SCR: `npx reqlan parse reqs/auth.rq` clean. Mention `parse reqs/ --fail-on-diagnostics` as the CI gate pattern. VO: Parse is how CI fails the build when requirements stop parsing — same sources the extension indexes. SCR: Ten-second export beat: `npx reqlan export --out ./docs-site` — same pipeline as the export form per [html_export]. VO: Export is one [commands] verb among equals. Form for exploration; CLI for scripts and pipelines. SCR: Slide: `--cwd` / `REQLAN_WORKSPACE`, `--json`, one active base per run. Explorer flash on nested-base dashed boundary — do not create one live. VO: Pin the workspace when the shell is in the wrong folder. Prefer `--json` for machines. Nested bases are separate runs — same craft as context and export in Advanced. SCR: Misconception beat — running from a parent monorepo without `--cwd`; results look empty or wrong. VO: If CLI and chat disagree, check cwd and active base first — not two different graphs. VO: Try this: from the demo root, run `npx reqlan parse reqs/ --fail-on-diagnostics` and imagine that line in CI. SCR: End card → Advanced overview. CTA from [phonebook].			Audience: viewers who finished the extension tour and want the headless twin named — CI, scripts, and agents without living in the editor. Outcome: install and run `reqlan` / `rq`; [init] a [base]; run search, analyse, and parse; know `--cwd` / `--json` and one-base-per-run; know export shares [html_export] with the extension. Runtime: 9 – 11 min. Demo: advanced get-started end state of `reqlan-tutorials-demo` — `.reqlan` indexed, `reqs/product.rq` + `reqs/auth.rq` present; integrated terminal at workspace root; extension idle in background ( not the focus ). Related: [cli_package] [commands] [init] [html_export] [base] [multi_base_environment] [llm_first] [function_parity]
gs_01_why_reqlan	SCR: Dark title card: reqlan wordmark + "Get started · 01 · Why requirements as code". Brand palette. VO: If you work with language models on real projects, you already write requirements — you just write them into the chat window, over and over. SCR: Split screen. Left: a bloated chat paste — one long paragraph mixing login, sessions, and audit rules. Right: three short named lines from the example. VO: reqlan is a small language for turning that intent into a graph of named ideas — so you, your teammates, and your tools can find, link, and reuse them instead of retyping. SCR: Brief montage (no narration overlap): VS Code with a `.rq` file, reqlan activity bar icon, Ideas Summary graph with a handful of nodes. VO: The day-one job is simple: name the idea, write the prose, point at related work. You can see it here — a file of names, and lines that connect them. SCR: Show a sticky note or chat bubble: "Isn't this just another notes app?" Cut to the named-ideas side of the contrast — three handles you can reference by name. VO: Notes apps store paragraphs. reqlan stores handles. You do not need a schema on day one — you need names your tools can search and link. SCR: Browser: Visual Studio Marketplace page for the reqlan extension — or Open VSX equivalent. Click Install. Switch to VS Code: Extensions view confirms install. File → Open Folder → create or select `reqlan-tutorials-demo`. VO: Install the reqlan extension from the Marketplace or Open VSX. Open a folder — even an empty one — so the editor has a workspace root. SCR: Onboarding webview appears: "Thanks for installing". Pause on docs / marketplace links in the welcome body. Do not create a base yet. VO: On first launch you get a short welcome screen. Keep those links handy. Creating a base — the `.reqlan` marker — comes after you have written a few ideas. SCR: End card: CTA row — Install · Docs · next episode "Your first idea". VO: Next: write your first idea in under a minute.			Audience: developers or knowledge workers who already talk to LLMs about their project and feel context drift. Outcome: viewer can explain reqlan in one sentence and knows where install / docs live. Runtime: 4 – 5 min. Demo: cold open on a fresh machine or reset profile. Folder `reqlan-tutorials-demo` does not exist yet — or an empty folder with only `README.md`. No `.rq` files. reqlan extension not installed ( or disabled ) until the install beat. Related: [mission_statement] [llm_first] [inspiration] [installation_event] [installation] [extension_installation]
gs_02_first_idea	SCR: Explorer in `reqlan-tutorials-demo`. Right-click → New Folder `reqs`. New File `reqs/product.rq`. Caret at line 1. VO: A reqlan file is just text with a `.rq` suffix. Start with the smallest form: a name, then the thought. SCR: Type keystroke-by-keystroke: `product_name reqlan tutorials demo`. Pause so token colours settle — name vs body. VO: That one line is already an idea. The name is a handle you will reference later. When you need more room, open a block. SCR: New line. Type the `welcome_flow` block from the example — open brace, body prose, close brace. Soft zoom on the body text inside braces. VO: Inside the braces, the first free text is the main description. No required metadata yet — just a name and the thought. SCR: Status bar: language mode shows reqlan (or Reqlan). Problems panel empty. Brief hover on bracket/brace highlighting if visible. VO: Save the file. If the extension is active, language support arrives immediately — highlighting, not a separate compile step. SCR: Chat bubble or sticky: "Do I need YAML front matter or a schema?" Cut back to the two-line file — no `@` attributes, no imports. VO: You do not. One-liners and blocks are valid on their own. Attributes and imports come later when you feel the need — not before your first idea exists. VO: Try this: rename `welcome_flow` to something that matches your own product, and rewrite the sentence in your voice. Save again. SCR: End card → next: "Link ideas".			Audience: freshly installed users who finished gs_01 or equivalent. Outcome: create a `.rq` file; write a one-liner and a block idea; save; see syntax highlighting and language mode. Runtime: 5 min. Demo: folder `reqlan-tutorials-demo` open in VS Code. Extension installed; onboarding seen once. Explorer shows empty tree or `README.md` only — no `reqs/` folder yet, no `.reqlan`. Related: [idea] [simple_idea] [block_idea] [simplicity] [file]
gs_03_link_ideas	SCR: Open `reqs/product.rq`. `welcome_flow` and `product_name` visible. Caret in the welcome_flow body. VO: Ideas earn power when they point at each other. Square brackets name a neighbour in the graph. SCR: Below the welcome prose, add a new block `onboarding_checklist` from the example. Inside `welcome_flow`, type `Success continues into [onboarding_checklist].` VO: The bracket link creates an edge. The target idea can live in the same file or another — the syntax is the same. SCR: F12 or Ctrl/Cmd-click `[onboarding_checklist]` — jump to the new block. Then Peek Definition (Alt+F12) back to the call site. VO: Follow the link — you are navigating a graph, not scrolling a long document. Peek when you want context without losing your place. SCR: In `onboarding_checklist`, add the wikilink: `Prefer [[welcome_flow\|the welcome screen]] as the way back if they bail early.` VO: Double-bracket wikilinks let you show a readable label right in the prose. The alias is for reading; the target is still the idea name `welcome_flow`. SCR: Sticky note: "Are wikilinks a different graph?" Show both link forms resolving to the same idea in go-to-definition. VO: Same graph, two spellings. Brackets for a quick hop; wikilinks when you want a readable label in prose. SCR: Optional five seconds: Command Palette → Reqlan: Show Local Graph if discoverable; otherwise stay in editor. VO: Try this: add a third idea `account_model` — accounts need an email and display name — and link both `welcome_flow` and `onboarding_checklist` to it once each. SCR: End card → next: "Your first base".			Audience: users with at least one idea file from gs_02. Outcome: use bracket references and a wikilink; peek definition and go-to-definition; understand ideas as a navigable graph. Runtime: 6 min. Demo: `reqlan-tutorials-demo/reqs/product.rq` open. Caret inside `welcome_flow`. File contains `product_name` one-liner and `welcome_flow` block only — no links yet, no `account_model`, no `.reqlan`. Related: [reference] [reference_brackets] [reference_wikilink] [idea]
gs_04_first_base	SCR: Activity bar reqlan icon — empty state CTA "Create base", or Command Palette: `Reqlan: Create Base`. Explorer: `reqs/product.rq` present, no `.reqlan` yet. VO: Installing the extension puts reqlan in the editor. It does not create a base. A base is a folder that owns a `.reqlan` directory — the boundary of one ideas graph. SCR: Sticky: "I installed the extension — why is the activity bar empty?" Point at missing `.reqlan` in Explorer. VO: Extension install and base create are two layers. Until the folder has a marker, there is no index boundary for neighbourhood views to bind to. SCR: Accept create-base at the workspace-folder root (`reqlan-tutorials-demo`). Explorer refreshes — expand `.reqlan`: `config.json`, `.rqignore`. Do not open the sqlite file. VO: That marker is shared application memory. Config and ignore rules live here. The ideas index appears under `.reqlan` once indexing activates — treat it as tooling memory, not a document you edit. SCR: Glance at activity bar or status: idea counts / readiness updating. Return to `product.rq` — caret on `welcome_flow`. VO: Same ideas you already wrote — now the tooling knows which graph they belong to. SCR: Optional ten seconds: terminal in a scratch empty folder — `npx reqlan init` — show identical `.reqlan` tree, then close terminal and return to demo folder. VO: Editor create-base and CLI init seed the same marker. Pick the host that fits the moment. SCR: Whiteboard beat — dashed box labelled "nested base" inside a parent folder. Do not create one live. VO: One base is enough for day one. Nested bases exist for larger trees — we name them later in Concepts. VO: Try this: confirm `.reqlan` is visible in Explorer, open `product.rq`, leave the caret on an idea, and watch the activity bar populate. SCR: End card → next: "Neighbourhood in the activity bar".			Audience: users with a small linked graph who have not yet marked a base. Outcome: create a [base] ( `.reqlan` ); explain that a base is the boundary of one ideas graph and owns the ideas index; know extension install ≠ base create. Runtime: 5 – 6 min. Demo: `reqlan-tutorials-demo` with `reqs/product.rq` containing linked `welcome_flow` ↔ `onboarding_checklist` ( and optional `account_model` ). No `.reqlan` folder. Activity bar or Ideas Summary shows empty / create-base prompt. Caret on any idea in `product.rq`. Related: [base] [base_detail] [create_base_onboarding] [base_installation] [installation] [extension_installation]
gs_05_activity_bar	SCR: Click the reqlan activity bar icon. Place caret inside `welcome_flow` in `reqs/product.rq`. Confirm active base label shows `reqlan-tutorials-demo` if a picker is visible. VO: The activity bar is your neighbourhood view — what sits near the idea under the cursor inside the active base — not a dump of the whole workspace. That keeps things readable when you copy a slice into chat later. SCR: Toggle "Sync with editor" on if off. Move caret from `welcome_flow` to `onboarding_checklist` — panes update. Briefly show hop depth control: 1 hop → 2 hops — neighbour list widens then tighten back to 1. VO: Sync follows your caret. Hop depth widens or tightens how far references travel. Keep it small while you are learning. Context binds to the active base's index — not a merged global graph. SCR: Sticky: "Should I open Ideas Summary every time?" Shake head — stay on activity bar. VO: No. The activity bar is for local context while you edit. Ideas Summary is for when you choose a broader map — not a habit on every keystroke. SCR: Activity bar header → Copy as Markdown (or equivalent action). Paste into an untitled scratch file — show compact markdown, not the whole `reqs/` tree. VO: Copy a focused slice when you want to talk to a model. Copying the neighbourhood keeps the answer sharp — pasting the whole requirements folder just buries the question. SCR: Command Palette → `Reqlan: Ideas Summary`. Graph or overview tab — centre on current idea, pan once, close tab or switch back to Explorer. Total on screen: under fifteen seconds. VO: Ideas Summary is for broader navigation when you choose it. Day one: glance and leave unless you are lost. VO: Try this: turn sync off, pin `welcome_flow`, click around the editor to other ideas, then turn sync back on and watch focus return. SCR: End card → next: "Ask @reqlan".			Audience: users with a small linked graph inside a [base]. Outcome: open the reqlan activity bar; follow editor focus with sync; adjust hop depth; copy a slice as markdown; glance at Ideas Summary without treating it as a full dump. Runtime: 6 – 7 min. Demo: `reqlan-tutorials-demo` is a [base] — `.reqlan/` present and index active. `reqs/product.rq` with linked `welcome_flow`, `onboarding_checklist`, optional `account_model`. Caret inside `welcome_flow`. Activity bar closed or on a non-reqlan view. Related: [activity_bar] [local_graph_view] [ideas_summary] [llm_first] [implied_context] [base]
gs_06_chat_search	SCR: Open `reqs/product.rq`. If `password_rules` is absent, add the block from the example and save. Open VS Code Chat (or Cursor Chat if recording host). VO: reqlan meets AI through focused tools — search, scoped context, small builds — not by shoving the whole graph into the prompt. SCR: VS Code Chat: type `@reqlan /rq-search password empty`. Show compact results — idea name, file path, one-line summary. Cursor alternate: invoke skill `/rq-search password empty` from chat. VO: Slash commands and skills use the `rq-` prefix. Search returns a short list on purpose — enough to pick, not enough to drown the thread. SCR: Sticky: "Why not paste product.rq?" Show result list vs a full-file paste thumbnail — reject the paste. VO: Search finds the handle. Add-to-context ships the slice. Meaning search beats scrolling when the word lives in body prose, not the idea name. SCR: Run add-to-context on `password_rules` — chat participant menu or `@reqlan` add action. Show only that idea's body joined to the thread, not the whole file. VO: Add what you need. Leave the rest in the index. SCR: Optional fifteen seconds: `@reqlan /rq-build-requirement` or Cursor skill `rq-build-requirement` with intent "rate-limit failed logins" — accept draft appended to `product.rq` as a new named idea; edit one line on camera. VO: Building a requirement from intent should land as a named idea you keep owning in source control — the model proposes; the `.rq` file remains the source of truth. VO: Try this: search for a word that only appears in a body sentence — like "twelve" inside `password_rules` — not in any idea name. SCR: End card → next: "Point ideas at code".			Audience: users ready to involve AI without abandoning the graph. Outcome: invoke `@reqlan /rq-search` ( VS Code Chat ) or Cursor skill `rq-search`; find an idea by meaning; add to context; optionally build a requirement from intent. Runtime: 6 min. Demo: same base and `reqs/product.rq` as gs_05. Add `password_rules` if missing — otherwise type it live at cold open. No chat thread open yet. `.reqlan` index active. Related: [ai_integration] [chat_skill_naming] [ai_add_to_context] [ai_build_requirement] [mcp_tools] [command_palette]
gs_07_link_code	SCR: New file `src/login.ts` — paste the stub from the example. New file `reqs/auth.rq` — type `login_api` block with file reference `Implementation: ["./src/login.ts"]` and link to `[password_rules]`. VO: Requirements earn their keep when they touch reality. Point an idea at a file with a file reference — the path lives in brackets like any other link. SCR: Ctrl/Cmd-click `["./src/login.ts"]` inside `login_api` — land in `login.ts`. Brief pause on the opened tab. VO: Same navigation habit as idea links — the graph knows about workspace files, not only `.rq` sources. SCR: Sticky: "File refs only go one way?" Place caret inside `login.ts` on the empty-password guard. Open activity bar — show requirements referencing this file or neighbourhood reacting to non-`.rq` focus. VO: Context scope works on any workspace file, not only `.rq` sources. Stand in the implementation and related requirements should still surface — the reverse path. SCR: Optional five seconds: mention folder references like `["./media/interviews/"]` on a slide — do not create live. VO: Folder references work the same way when assets span a directory. SCR: Recap montage (five seconds): seven episode titles stacked — Why · First idea · Link · Base · Activity bar · Ask · Point at code. VO: You now have the get-started loop: name ideas, link them, create a base, inspect the neighbourhood, ask the tools with a small slice, and stitch requirements to code. VO: Try this before you go: link one more source file your welcome flow actually depends on — a component, route, or test. SCR: End card: "Concepts · The building blocks, the extension map, then the CLI" — series handoff to [concepts_series].			Audience: users with application source beside `reqs/` who completed gs_01 – 06. Outcome: add a file reference from an idea to implementation; open the file from the requirement; see reverse scope when standing in code; recap the get-started loop into Concepts. Runtime: 6 – 7 min. Demo: base active. `reqs/product.rq` with linked ideas and `password_rules`. No `reqs/auth.rq` yet. No `src/` folder. Caret in `product.rq`. Explorer shows only `reqs/` and `.reqlan/`. Related: [any_file_scope] [context_scope] [reference] [referenced_files] [file]