3GPP Analyzer

3GPP Analyzer (3gppanalyzer.com) is an AI-powered 3GPP specification database for RAN and telecom engineers — browse, search, and understand LTE and 5G technical specifications with version tracking, release timelines, technology guides, glossary terms, and Claude-powered AI summaries grounded in official Scope excerpts.

Live site: https://3gppanalyzer.com

Stack

Frontend: Next.js 14 (App Router) + Tailwind CSS (light/dark theme), deployed on Vercel
API: Cloudflare Worker (3gpp-analyzer-worker)
Database: Cloudflare D1 (SQLite at the edge)
Object storage (optional / planned): Cloudflare R2
AI summaries: Anthropic Claude API (claude-sonnet-4-6), grounded in official Scope text when available

Note: Product branding is 3GPP Analyzer / 3gppanalyzer.com. Cloudflare D1/R2 resource names still use the legacy 3gpp-sniffer-db / 3gpp-sniffer-specs identifiers so existing production data stays bound without a rename migration. Account-specific IDs and Worker URLs live only in gitignored wrangler.local.toml / hosting dashboards.

Catalog Coverage

Live ingest covers Rel-15–Rel-20 LTE and 5G series from the public FTP archive. Discontinued 2G/3G (GERAN/UTRAN) series are excluded from the catalog and ingest. Featured technology areas include 5G-Advanced, NTN, Network Slicing, 5G NR, and LTE evolution tracks.

AI summaries prefer official Scope excerpts extracted from DOCX packages; specs without an excerpt show a short catalog-metadata note instead of invented content.

Local Development

Requires Node.js 20.x (see .nvmrc / package.json engines). Wrangler is included as a dev dependency.

npm install

# Local Cloudflare config (gitignored) — copy the template and set your D1 database_id
cp wrangler.toml wrangler.local.toml
# Edit wrangler.local.toml: replace <D1_DATABASE_ID> with the id from the Cloudflare dashboard

# 1. Create the local D1 database and load schema + seed data
npm run db:schema
npm run db:seed

# Optional: apply incremental migrations if the DB already exists
# (DB name `3gpp-sniffer-db` is the legacy Cloudflare resource id — see note above)
npx wrangler d1 execute 3gpp-sniffer-db --local --file=./migrations/0001_ai_rate_limits.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --local --file=./migrations/0002_source_excerpts.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --local --file=./migrations/0003_catalog_accuracy.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --local --file=./migrations/0004_ntn_and_retire_2g3g.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --local --file=./migrations/0005_network_slicing.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --local --file=./migrations/0006_purge_2g3g.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --local --file=./migrations/0007_technology_icons.sql --yes

# 2. Start the Worker API (port 8787)
npm run worker:dev

# 3. In a second terminal, start the Next.js frontend (port 3000)
npm run dev

Visit http://localhost:3000.

Environment

Variable	Where	Purpose
`ANTHROPIC_API_KEY`	`.dev.vars` (local) / Worker secret (prod)	AI summary generation
`NEXT_PUBLIC_API_URL`	Next.js / Vercel env	Worker API base URL (default `http://localhost:8787`)
`NEXT_PUBLIC_SITE_URL`	Next.js / Vercel env	Canonical site URL for SEO, sitemap, Open Graph

Copy .dev.vars.example → .dev.vars and set ANTHROPIC_API_KEY for local AI summaries.
Copy .env.example → .env.local for frontend public env vars.

# Production Worker secret (never put this in Vercel)
wrangler secret put ANTHROPIC_API_KEY

Vercel (frontend only) — set these project env vars for Production/Preview:

Name	Value
`NEXT_PUBLIC_API_URL`	Your deployed Worker origin (prefer `https://api.3gppanalyzer.com` once the custom domain is live; otherwise the Worker’s `*.workers.dev` URL from the Cloudflare dashboard)
`NEXT_PUBLIC_SITE_URL`	`https://3gppanalyzer.com`

Do not put ANTHROPIC_API_KEY in Vercel — that secret stays on the Cloudflare Worker (wrangler secret put ANTHROPIC_API_KEY).

Cloudflare domain setup (`3gppanalyzer.com`)

The Worker is deployed as 3gpp-analyzer-worker. To serve it on api.3gppanalyzer.com:

In Cloudflare Dashboard → Add a site → enter 3gppanalyzer.com
At your domain registrar, switch nameservers to the Cloudflare NS values shown
Wait until the zone status is Active
In wrangler.toml, uncomment the [[routes]] block for api.3gppanalyzer.com
Run npm run worker:deploy
Point Vercel NEXT_PUBLIC_API_URL at https://api.3gppanalyzer.com

Until the zone is on Cloudflare, point Vercel at your Worker origin from the Cloudflare dashboard. Do not commit account-specific Worker URLs or resource IDs to this repository.

Deployment

# One-time: create the D1 database, then set database_id in a local wrangler override
# (see wrangler.toml comments — do not commit account-specific IDs)
npm run db:create

# Push schema + seed data to the remote database
npm run db:schema:remote
npm run db:seed:remote

# Apply incremental migrations on remote D1 (legacy resource name `3gpp-sniffer-db`)
npx wrangler d1 execute 3gpp-sniffer-db --remote --file=./migrations/0001_ai_rate_limits.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --remote --file=./migrations/0002_source_excerpts.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --remote --file=./migrations/0003_catalog_accuracy.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --remote --file=./migrations/0004_ntn_and_retire_2g3g.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --remote --file=./migrations/0005_network_slicing.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --remote --file=./migrations/0006_purge_2g3g.sql --yes
npx wrangler d1 execute 3gpp-sniffer-db --remote --file=./migrations/0007_technology_icons.sql --yes

# Deploy the Cloudflare Worker (3gpp-analyzer-worker)
npm run worker:deploy

# Build / deploy the Next.js app on Vercel (3gppanalyzer.com)
# Set NEXT_PUBLIC_API_URL to your Worker origin (api.3gppanalyzer.com or *.workers.dev)
# Set NEXT_PUBLIC_SITE_URL to https://3gppanalyzer.com
npm run build

SEO

Canonical host: https://3gppanalyzer.com
Root metadata, Open Graph, and Twitter cards in app/layout.tsx
JSON-LD (WebSite + SoftwareApplication) sitewide; TechArticle on spec pages
Canonical URLs via NEXT_PUBLIC_SITE_URL (lib/seo.ts)
Spec pages use clean paths: /spec/{specId}/{release}/ (legacy ?specNumber= redirects)
app/sitemap.ts — home, listings, featured tech pages (NTN, Network Slicing, 5G-Advanced, …), releases, glossary terms, and every versioned spec
app/robots.ts — allows crawl and points to https://3gppanalyzer.com/sitemap.xml
Per-route metadata / generateMetadata on major pages

Data Ingestion

scripts/seed.sql loads key specifications, releases, technology areas, and glossary terms so the site works out of the box.

Crawl specs from the public 3GPP FTP archive:

# Default selected LTE/5G series across Rel-15–Rel-19
npm run ingest

# Discover published series for a release (GERAN/UTRAN series are skipped)
node scripts/ingest.js --releases=Rel-15,Rel-16,Rel-17,Rel-18,Rel-19 --all-series
npm run ingest:rel20

Progress is checkpointed in scripts/.ingest-checkpoint.json (gitignored) and requests are rate-limited.

After a bulk ingest, refresh taxonomy / stub titles and push SQL to remote D1:

npm run backfill:metadata
node scripts/upload-sql-chunks.js scripts/backfill-out.sql 100

node scripts/export-local-specs.js --release=Rel-19 --out=scripts/ingest-out.sql
node scripts/upload-sql-chunks.js scripts/ingest-out.sql 60

Extract short official Scope excerpts from DOCX archives for source-grounded summaries:

npm run extract:sources
node scripts/upload-sql-chunks.js scripts/source-excerpts.sql 40

AI endpoint behavior

Cached summaries: readable via GET or POST
New grounded generations: require Anthropic key + Scope excerpt; rate-limited (3 uncached gens / IP / hour, 100 / day globally)
Specs without a Scope excerpt: short deterministic metadata note (no Anthropic call)

Project Structure

Path	Purpose
`app/`	Next.js App Router pages, sitemap, robots
`components/`	Shared UI (nav, cards, filters, theme toggle, AI summary)
`content/`	Hand-written technology guide content
`lib/`	API client, SEO helpers, Anthropic client, coverage labels
`migrations/`	Incremental D1 migrations
`worker/`	Cloudflare Worker API routes
`scripts/`	Seed SQL, FTP ingest, Scope extraction, D1 upload helpers
`public/`	Static assets (logo marks)

License

Source code is released under the MIT License.

The license covers this project's code only. Specification content is the property of 3GPP and remains subject to 3GPP's own terms of use.

Disclaimer

3GPP Analyzer data is sourced from the 3GPP FTP server. This is an unofficial community tool, not affiliated with 3GPP or any mobile operator.

Name		Name	Last commit message	Last commit date
Latest commit History 26 Commits
app		app
components		components
content		content
lib		lib
migrations		migrations
public		public
scripts		scripts
worker		worker
.dev.vars.example		.dev.vars.example
.env.example		.env.example
.gitignore		.gitignore
.nvmrc		.nvmrc
.vercelignore		.vercelignore
LICENSE		LICENSE
README.md		README.md
next-env.d.ts		next-env.d.ts
next.config.js		next.config.js
package-lock.json		package-lock.json
package.json		package.json
postcss.config.js		postcss.config.js
schema.sql		schema.sql
tailwind.config.ts		tailwind.config.ts
tsconfig.json		tsconfig.json
tsconfig.worker.json		tsconfig.worker.json
vercel.json		vercel.json
wrangler.toml		wrangler.toml

Navigation Menu

Repository files navigation

3GPP Analyzer

Stack

Catalog Coverage

Local Development

Environment

Cloudflare domain setup (`3gppanalyzer.com`)

Deployment

SEO

Data Ingestion

AI endpoint behavior

Project Structure

License

Disclaimer

About

Releases

Packages

Contributors

Languages

Folders and files

Latest commit

History

Repository files navigation

3GPP Analyzer

Stack

Catalog Coverage

Local Development

Environment

Cloudflare domain setup (3gppanalyzer.com)

Deployment

SEO

Data Ingestion

AI endpoint behavior

Project Structure

License

Disclaimer

About

Topics

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages

Cloudflare domain setup (`3gppanalyzer.com`)