The Release Ops Command Center
Last reviewed
AdvancedWhat you'll learn
~35 min- Build a Node.js static site generator producing a multi-page command center
- Understand the SSG pattern: data + templates = deployable HTML
- Integrate outputs from Lessons 1-5 into a unified dashboard
- Deploy the command center behind a login on Netlify or Vercel
What you’re building
You’ve built five tools across this module. Asset engine. Campaign generator. YouTube episode builder. Analytics dashboard. Royalty reconciler. They all work. They all produce useful output. But right now, the output lives in five different folders, and every week you’re opening files in three different places to figure out what to do next.
You’re going to build one command center that pulls everything together. A static website — generated from your data files — that gives you a single place to see this week’s priorities, your campaign status, your revenue numbers, and your platform analytics. Not a live app with servers and databases. Just HTML files generated from your data, deployed for free, accessible from your phone or laptop anywhere.
This isn’t a fan-facing website. It’s your operations dashboard, and it shows your revenue. Think of it like a cockpit: all the instruments in one place so you can make decisions without switching between five different tools. A URL nobody knows is not private: anyone who has it can read every number. Deploy it behind a login (the deploy section below shows how on Netlify and Vercel) or just open the built files on your own machine.
How the pieces connect
Here’s what you built in each lesson and what data flows into the command center:
| Lesson | Tool | Output the command center uses |
|---|---|---|
| L1 | Release Asset Engine | Asset file paths (generated artwork) |
| L2 | Campaign Generator | campaign.json (built from the captions and calendar L2 writes) |
| L3 | Episode Builder | Episode metadata (next YouTube publish date) |
| L4 | Control Room | analytics.csv (built from the Spotify/Shopify exports L4 reads) |
| L5 | Royalty Reconciler | royalties.csv (built from the distributor CSVs and flags L5 reads and reports) |
The command center reads from a release-data/ directory. You drop your data files there. One command generates the whole site.
Data files + templates = a deployable website. That’s it. Astro (this very training site), Hugo, Jekyll, Eleventy — they all work this way. You’re building a tiny version of the same pattern. No server runs your site. No database stores your data. A build script reads files and writes HTML. You deploy the HTML. Done.
Before you start
First, if your AI tool is still open, leave it: type /exit and press Enter (if you use Antigravity, press Ctrl+D with nothing typed instead), and wait until you see the $ or % prompt. Make a clean folder for this lesson inside ~/producer-workbench:
mkdir -p ~/producer-workbench/06-release-ops-command-center && cd ~/producer-workbench/06-release-ops-command-centerOn Windows PowerShell, run mkdir -Force ~/producer-workbench/06-release-ops-command-center; cd ~/producer-workbench/06-release-ops-command-center, or open the folder in your desktop agent app. Then start your agent the way your track’s setup lesson showed.
Switch to plan mode
Once your agent is open, switch to Plan mode inside the session (Shift+Tab, /plan, or your desktop app’s mode picker). In plan mode it asks clarifying questions and shows you a plan before building.
The conversation
Each message below is an example. Say it in your own words.
Say what you want
Start with a short plain-English description of what you want to build:
I want to build a static site generator that reads release, campaign, analytics, and royalty files from release-data/ and builds a four-page dark-theme command center into dist/. Ask me what you need first.Answer its questions
Your agent will ask a few questions about data inputs, pages, and deployment. Answer each briefly:
| If it asks | You might say |
|---|---|
| Which input files should the generator read? | Read release-info.json, campaign.json, analytics.csv, and royalties.csv from a release-data/ folder using standard command-line tools. |
| What four HTML pages should the build script generate in dist/? | Build index.html (This Week’s Actions), campaign.html (Campaign Status), revenue.html (Revenue Dashboard), and analytics.html (Platform Analytics). |
| What should happen if one of the data files is missing? | Log a warning and fall back to empty defaults so the site still builds with partial data. |
| What visual theme and navigation rules should the pages follow? | Dark #09090b theme with #f97316 orange accents, relative sibling nav links without leading slashes, and responsive cards. |
| Should we include sample data and deploy configs? | Yes, include realistic sample files for JADED in release-data/, plus netlify.toml, vercel.json, and a deploy:github-pages script. |
Read the plan
Before approving, check the plan for four domain essentials:
- All internal links are sibling-relative. Navigation links and stylesheet links use
campaign.html,revenue.html, andstyle.css(without a leading/) so the pages work both when opened directly fromdist/index.htmlon disk and when hosted in a folder. - Missing data files degrade gracefully. If
analytics.csvorroyalties.csvis absent,data-loader.jslogs a warning and supplies empty defaults rather than crashing the build. - This Week’s Actions synthesizes priorities across all four files.
index.htmlhighlights the next scheduled social post, any flagged royalty rows, release countdowns, and streaming or revenue milestones. - Sample data and deployment configs work out of the box. Running
node src/build.jsreads the includedrelease-data/sample files (including 14 campaign posts and 2 flagged royalty rows) and outputsdist/ready fornetlify.tomlorvercel.json.
If the first plan uses root-relative /campaign.html links or crashes on missing CSVs, say so briefly:
Use relative nav and CSS links with no leading slash so dist/index.html works from disk, and use empty defaults if a data file is missing.When the plan covers all four, approve it: pick the option to proceed or say Yes, go ahead and build it (in Codex, press Shift+Tab first to leave plan mode). Once you verify node src/build.js generates all four pages in dist/, ask your agent to save a Git checkpoint:
Save a checkpoint with Git so we can roll back if needed.What you get
After your AI CLI tool finishes, you’ll have a project folder structure like this:
REFERENCE PROJECT LAYOUT command-center/ generated static site generator folder tree open reference hide reference Reference example of the folder layout your agent creates. Open to see where templates, static assets, and release-data/ inputs live.
command-center/ package.json netlify.toml vercel.json src/ build.js data-loader.js templates/ layout.hbs index.hbs campaign.hbs revenue.hbs analytics.hbs static/ style.css release-data/ release-info.json campaign.json analytics.csv royalties.csvBuild and preview it
Ask your agent to install what is needed, build the command center, and open dist/index.html in your browser:
Install what the project needs, build the command center site, and open dist/index.html in my browser.If you prefer running the build in a terminal yourself:
cd command-centernpm installnode src/build.jsYou should see output like:
Loading data files... ✓ release-info.json (1 release) ✓ campaign.json (14 posts) ✓ analytics.csv (127 rows) ✓ royalties.csv (24 rows, 2 discrepancies)
Building pages... ✓ dist/index.html -- This Week's Actions ✓ dist/campaign.html -- Campaign Status ✓ dist/revenue.html -- Revenue Dashboard ✓ dist/analytics.html -- Platform Analytics ✓ dist/style.css
Build complete: 4 pages generatedOpen dist/index.html in your browser. Click through all four pages. Check:
- Does the navigation work between pages?
- Does “This Week’s Actions” show relevant priorities from the sample data?
- Does the revenue page show discrepancy flags?
- Is the dark theme applied consistently?
While tweaking templates, you can add a watch mode: "Add a watch mode that automatically rebuilds the site whenever any template or data file changes." This makes it easy to iterate on the design without rerunning the build manually.
A worked example: building for a new single
Here’s a real scenario. You’re releasing “JADED” on March 13. Your data:
- release-info.json — You fill in the track name, release date, and your top priorities for the week
- campaign.json: Built from the Campaign Generator’s output (Lesson 2), which writes Markdown files such as
captions.mdandcalendar.md, not JSON. Ask your AI CLI tool to turn them into the campaign.json shape the command center’s sample data uses. - analytics.csv: Built from the same Spotify for Artists and Shopify CSV exports you loaded into the Control Room (Lesson 4). Ask your AI CLI tool to merge them into the columns the command center expects.
- royalties.csv: Built from the same distributor CSVs the Royalty Reconciler (Lesson 5) reads; its output is an HTML report, not a CSV. Ask your AI CLI tool to merge them into the royalties.csv columns the command center expects, with the reconciler’s discrepancy flags as a column.
Drop all four files into release-data/, run the build, and open the site. Your “This Week’s Actions” page might show:
- URGENT: Post Instagram teaser caption (scheduled for tomorrow, 6pm ET)
- URGENT: Follow up on MOOD’s Apple Music royalty discrepancy (-18% streams)
- NORMAL: Release day is in 4 days — EPK is ready, check all platform links
- INFO: Spotify listeners up 8% this week (52,400 → 56,590)
- INFO: KICK IT W/U crossed $1,000 total revenue this quarter
That’s your week in one glance. No digging through five different folders.
Deploying your command center
The generated site is just HTML and CSS. No server needed. Netlify and Vercel can both host it free behind a login; GitHub Pages is free but public.
Whichever host you pick, the first deploy goes live at once, so make it from the sample data the prompt generated. Turn the login on, check the URL from a private browser window (it should ask you to sign in), and only then rebuild with your real files and deploy again.
Netlify
# Install Netlify CLI (one time)npm install -g netlify-cli
# Deploycd command-centernetlify deploy --prod --dir=distThen make the project private in its Netlify settings (project visibility). On the Free plan, only you (the team owner) can open it.
Vercel
# Install Vercel CLI (one time)npm install -g vercel
# Deploycd command-centervercel --prodThen turn on Vercel Authentication, which is free on every plan (Vercel’s changelog): open the project in the Vercel dashboard, go to Deployment Protection, switch Vercel Authentication on, select the environments to protect (include production, since vercel --prod deploys there), and select Save (how-to). The site then asks for your Vercel login. Vercel’s free Hobby plan is for personal, non-commercial use, so check its terms if the dashboard is part of your business.
GitHub Pages
# Build and copy to docs/npm run deploy:github-pages
# Commit and pushgit add docs/git commit -m "Update command center"git pushThen enable GitHub Pages in your repo settings, pointing to the docs/ folder. GitHub Pages sites are public, even from a private repository, and on GitHub Free the repository itself must be public (GitHub Pages docs). Use this option only for a version of the site with no revenue or royalty data.
🔍Why a static site instead of a 'real' appclick to expand
You might be wondering: why not build a React app with a database? Or use a service like Notion or Airtable?
A static site wins for this use case because:
- No server costs: Netlify and Vercel both have free plans that host static files, with a login in front at no extra cost.
- No maintenance: There’s no database to back up, no server to keep running, no dependencies to update continuously. The site is just files.
- Fast: Static HTML loads instantly. No JavaScript frameworks, no API calls, no loading spinners.
- Nothing to query: Your data is baked into the HTML at build time, so there is no database to break into. But anyone who can open the URL can read every number on the page, which is why the deploy step puts it behind a login. The built site contains only the data you chose to include.
- Portable: You can open the HTML files directly from your filesystem without deploying anywhere. It works offline.
The tradeoff: you have to rebuild when data changes. For a weekly check-in dashboard, that’s running one command before you look at it. Perfectly fine.
Netlify and Vercel both host a 4-page static site on their free plans and can keep it behind your own login. GitHub Pages is free too, but every Pages site is public. If you’ve never deployed anything, Netlify Drop (app.netlify.com/drop) is the easiest start: drag in a dist folder built from the sample data (it goes live at once, before you sign in), then sign in, make the project private, and only then deploy your real numbers.
If something is off
| Problem | Follow-up prompt |
|---|---|
| Navigation links are broken between pages | The nav links use absolute paths (/campaign.html) but the site is deployed to a subdirectory. Change all nav links to relative paths (campaign.html, revenue.html, etc.) so they work whether the site is at the root or in a subdirectory. |
| Missing data file crashes the build | The build crashes when analytics.csv is missing. If a data file doesn't exist, log a warning, fall back to an empty default so the site still builds, and show "No data available" in affected sections. |
| The CSS isn’t applied to generated pages | The pages are unstyled. Make sure the build copies style.css into dist/ after creating the folder and links to style.css using a relative path with no leading slash. |
| Template placeholders show raw curly braces | The generated pages are showing raw placeholders like {{revenue}} instead of actual values. Make sure each template is rendered with the loaded data before writing the HTML files to dist/. |
Customize it
Add automatic rebuild on data change
Add a watch command that monitors release-data/ and templates/ and automatically rebuilds the site with a timestamp whenever a file changes.Add an email digest
Add a --digest flag to build.js that writes a plain-text email summary in dist/digest.txt with top actions, total revenue, flags, and upcoming posts.Add social media preview images
Add Open Graph meta tags to each page and generate a 1200x630 dark-theme PNG preview image for each page showing its title and key metric.Add historical release comparison
Support multiple release folders under release-data/ and generate a releases.html comparison page highlighting the top release by first-week streams, revenue, and merch conversion.When Things Go Wrong
Use the Symptom → Evidence → Request pattern: describe what you see, paste the error, then ask for a fix.
The story so far
Over six lessons, you’ve built the operations backbone:
- Release Asset Engine (Lesson 1) — one cover image in, six platform-sized assets out
- Campaign Generator (Lesson 2) — one metadata file in, captions, hashtags, calendar, and an EPK out
- Episode Builder (Lesson 3) — one notes file in, all YouTube publishing metadata out
- Control Room (Lesson 4) — CSV exports in, geographic and business intelligence out
- Royalty Reconciler (Lesson 5) — distributor CSVs in, discrepancy flags and revenue report out
- Command Center (Lesson 6) — all of the above in, one weekly action-plan dashboard out
These tools chain together. The campaign generator’s output feeds the command center’s campaign page. The reconciler’s flags appear on the revenue dashboard. The control room’s analytics populate the analytics page. The command center is the hub that ties every tool into a single view.
Next up: building your signature visualizer content — the tape deck aesthetic that makes your brand recognizable across every platform.
Try it yourself
- Open your AI tool in your lesson folder and switch to plan mode.
- Talk your agent through the plan and approve it to build the project.
- Ask your agent to install what is needed and build the site (or run
npm installandnode src/build.jsin your terminal). - Open
dist/index.htmland click through all four pages. - Replace the sample data with real output from your Lessons 1-5 tools.
- Deploy to Netlify or Vercel behind a login.
Key takeaways
- Static site generators are data + templates — no server, no database, no complexity. Read files, render templates, write HTML. That’s the entire pattern.
- Free deployment is real: Netlify and Vercel host static sites on free plans, behind a login at no extra cost. GitHub Pages is free too, but public. Your command center costs nothing to run.
- Integration is the capstone skill — building individual tools is valuable, but connecting them into a unified workflow is where the real power is. The command center doesn’t do anything new; it orchestrates what you already built.
- Private dashboards beat scattered files — one URL with all your release data is better than five folders you have to remember to check.
- Rebuild, don’t maintain — when data changes, rebuild the site. There’s no state to manage, no cache to invalidate, no database to migrate. Just run the build and deploy.
Your command center works locally but you want to deploy it. You choose Netlify. What does Netlify actually serve to visitors?
What’s next
The operations side is locked in. Next lesson: you build your signature audio visualizer — the tape deck aesthetic that turns every track into content for Reels, Shorts, and TikTok.