Octonet

Octonet

A multi-protocol API server that exposes a single ORM schema over 11 transports including MCP, enabling AI agents to interact with data from 13 different databases via a unified interface.

Category
Visit Server

README

@mostajs/net — Octonet

One schema, 11 transports, 13 databases, multi-tenant auth — out of the box. Schema-driven multi-protocol API server for the @mostajs ecosystem. Built with TypeScript on Fastify, designed for polyglot consumption (14 native NetClients).

<p align="center"> <img src="logo/octonet-icon.svg" width="128" alt="Octonet Logo"/> </p>

npm License mcp.so

Author : Dr Hamid MADANI <drmdh@msn.com> Homepage : octonet.amia.fr · mcp.amia.fr


<p align="center"> <img src="assets/architecture-orm-net-netclient.png" width="900" alt="Octonet stack as electrical sockets: @mostajs/orm plugs into 13 databases, @mostajs/net (Octonet) re-exposes them over 11 transports, Data-Plug bridges them, and NetClient fans out to 18 language clients consumed by any external app."/> </p>

The socket board. @mostajs/orm connects to 13 databases; Octonet (@mostajs/net) re-exposes the same entities over 11 transports (REST, WebSocket, gRPC, MQTT, AMQP…); NetClient then fans out to 18 language clients — so any external app plugs into your data, in any runtime.


Table of Contents

  1. What is Octonet
  2. Try in 2 clicks (T1 sandbox)
  3. 3-tier onboarding model
  4. 11 transports
  5. 13 databases
  6. Multi-tenant authentication
  7. Generic scope registry
  8. Multi-project support
  9. Quick start (self-host)
  10. Configuration (.env)
  11. API endpoints reference
  12. Architecture
  13. Boot sequence
  14. 14 polyglot NetClients
  15. Production deploy (amia.fr)
  16. License

What is Octonet

Octonet (@mostajs/net) is a Node.js multi-transport server that exposes a single @mostajs/orm schema as 11 different network protocols simultaneously : REST, GraphQL, WebSocket, SSE, JSON-RPC, MCP, gRPC, tRPC, OData, NATS, Arrow Flight.

The same User entity is reachable as :

curl https://octonet.amia.fr/api/v1/User              # REST
curl https://octonet.amia.fr/graphql -d '{"query":...}'  # GraphQL
wscat -c wss://octonet.amia.fr/ws                     # WebSocket
curl -N https://octonet.amia.fr/mcp                   # MCP (Claude Desktop)
# … 7 more

Backed by @mostajs/orm (13 SGBD dialects). Backed by @mostajs/rbac + @mostajs/auth + @mostajs/api-keys for multi-tenant identity. Backed by @mostajs/mproject for multi-project routing.

Octonet is the second cerebral lobe of the @mostajs trilogy :

  • #1 — @mostajs/orm : data persistence (13 databases)
  • #2 — Octonet : multi-protocol transport (11 wire protocols)
  • #3 — NetClients polyglottes : 14 native client libraries (Java, .NET, Python, Go, Swift, Kotlin, Dart, Rust, PHP, Ruby, Elixir, Lua, Delphi, Unity)

Try in 2 clicks (T1 sandbox)

The fastest way to test Octonet from any of the 14 runtimes :

  1. Open https://octonet.amia.fr/try
  2. Pick an alias (e.g. alice-42)
  3. Get an apikey scoped to your sandbox
curl -X POST https://octonet.amia.fr/try \
  -H "Content-Type: application/json" \
  -d '{"alias":"alice-42"}'

# Response:
{
  "status": "ok",
  "data": {
    "alias": "alice-42",
    "projectSlug": "sandbox-alice-42",
    "apiKey": "sk_test_…(64 chars, shown ONCE)",
    "permissions": { "projects":["sandbox-alice-42"], "operations":["read","write"], "transports":["rest","mcp"] },
    "expiresAt": "2026-05-02T18:38:19.920Z",
    "quota": { "reqPerDay": 500 },
    "exampleCurl": "curl https://octonet.amia.fr/api/v1/sandbox-alice-42/User -H \"X-API-Key: sk_test_…\"",
    "mcpUrl": "https://octonet.amia.fr/mcp"
  }
}

Your sandbox is :

  • A private SQLite file (isolated from other aliases)
  • Pre-seeded with User, Product, Order entities
  • Read+write CRUD via REST and MCP
  • 500 requests/day quota
  • TTL 7 days (auto-deleted if idle)
  • Free, no email required, anti-abuse rate-limited (10 sandboxes/h/IP)

Use the apikey in any of the 14 NetClients :

# Java / .NET / Python / Go / etc.
export MOSTAJS_NET_URL=https://octonet.amia.fr
export MOSTAJS_NET_API_KEY=sk_test_…

3-tier onboarding model

Tier Cible Auth Quota Use case
T1 — Sandbox publique /try "I want to test in 10 minutes" alias seul (pas d'email) 500 req/jour, TTL 7j démo NetClients, intégration tests
T2 — Compte enregistré octocloud.amia.fr "j'utilise vraiment, gratuit" email + mot de passe 10 000 req/jour apps personnelles, side-projects
T3 — Self-host npx @mostajs/net init "tes données chez toi" admin local aucune enterprise, on-premise, AGPL

11 transports

Auto-generated from your registered schemas. Toggle each via env var (default : all enabled).

# Transport Endpoint Use case Notes
1 REST /api/v1/{Entity} · /api/v1/{project}/{Entity} universel 15 routes par entité (CRUD + count + search + aggregate + bulk + relations)
2 GraphQL /graphql (POST) front-end riches schéma + GraphiQL IDE auto-générés via mercurius
3 WebSocket wss://…/ws temps réel events entity.created/updated/deleted/upserted broadcast
4 SSE /events (GET stream) mobile / browser-friendly server-sent events
5 JSON-RPC /rpc (POST) EVM-adjacent, classic JSON-RPC 2.0 + method discovery
6 MCP /mcp (POST/GET SSE) agents IA (Claude, ChatGPT) 15 tools/entité auto-générés (listed on mcp.so)
7 gRPC :50051 inter-services low-latency .proto auto-généré, 6 RPCs/entité
8 tRPC /trpc/{Entity}.{op} TypeScript fullstack type generation côté client
9 OData /odata/{Collection} (+ $metadata) SAP/Microsoft Dynamics OData v4 ($filter, $select, $orderby)
10 NATS mostajs.{Entity}.{op} pub/sub edge request-reply messaging
11 Arrow Flight /arrow/* analytics columnaire streaming zero-copy

Le même schéma, 11 portes d'entrée différentes, 0 codegen.


13 databases

Persistence dialects fournis par @mostajs/orm :

Catégorie Bases Dialect ID
SQL mainstream PostgreSQL, MySQL, MariaDB, SQLite postgres, mysql, mariadb, sqlite
SQL enterprise Oracle, SQL Server, DB2, SAP HANA, HSQLDB, Sybase oracle, mssql, db2, hana, hsqldb, sybase
NewSQL / Cloud CockroachDB, Google Cloud Spanner cockroachdb, spanner
NoSQL MongoDB mongodb

Switch dialect = changer 1 ligne d'env :

DB_DIALECT=postgres → DB_DIALECT=mongodb
SGBD_URI=postgresql://… → SGBD_URI=mongodb://…

Multi-tenant authentication

Octonet utilise un middleware d'authentification basé sur API keys avec scopes (orienté machine-to-machine) + RBAC pour la gestion humaine. L'orchestrateur agnostique vit dans @mostajs/auth/lib/check-request.ts.

Flux d'auth d'une requête

HTTP Request
  │
  ▼
authGuard (Fastify adapter)
  │
  ▼
checkRequest (@mostajs/auth — framework-agnostic)
  │
  ├─ extract X-API-Key from header / Bearer token / ?apikey=
  ├─ resolveApiKey (@mostajs/api-keys) — DB lookup + bcrypt verify
  ├─ isScopeAuthorized (@mostajs/api-keys) — check scope.values
  │     for each (scope, value) pair :
  │       checks: [{scope:'projects', value:slug},
  │                {scope:'operations', value:'read'|'write'|'admin'},
  │                {scope:'transports', value:'rest'|'mcp'|...}]
  ├─ touchApiKey — fire-and-forget : lastUsedAt, usageCount, lastIp
  │
  ▼
ormHandler (entité CRUD) ─→ sanitizer (strip password/hash/tokens)
  │
  ▼
HTTP Response

Scope-based permissions (generic, extensible)

Une apikey a la shape :

{
  permissions: {
    scopes: {
      projects:   ['my-project','demo'] | '*',     // declared by @mostajs/mproject
      operations: ['read','write','admin'] | '*',  // declared by @mostajs/orm
      transports: ['rest','graphql','mcp'] | '*',  // declared by @mostajs/net
      // anything: any module can register a new scope
    },
    rateLimit: 500,
  }
}

@mostajs/api-keys ne connaît PAS les noms projects/operations/transports — c'est volontaire. Chaque module enregistre ses scopes au boot via registerScope(dialect, {name, label, …}). Le check est générique : isScopeAuthorized(perms, scope, value).

HTTP status mapping

Cas HTTP code Body
Pas d'apikey 401 {error: {code:'UNAUTHORIZED', message:'API key required…'}}
Apikey invalide / révoquée 401 {error: {code:'UNAUTHORIZED', message:'Invalid or revoked API key'}}
Apikey valide hors scope 403 {error: {code:'FORBIDDEN', message:'API key not authorized for X="Y"'}}
Apikey OK 200/201 {status:'ok', data: …} (sanitized — no password/hash)

MCP fallback (compat mcp.so)

L'endpoint /mcp autorise un fallback automatique sur l'apikey labelée public-default quand aucune clé n'est présentée — préserve la compat avec mcp.so / Claude Desktop. La clé publique est read-only sur le projet default uniquement → écritures bloquées.

Sanitizer

Les champs password, hash, verifyToken, resetToken, apiKeyHash, secret, privateKey sont automatiquement strippés de toutes les réponses JSON par un middleware global. Aucun risque de leak à travers /api/v1/User?limit=1.


Generic scope registry

Le catalogue des scopes vit en base de données du projet accueillant (pas de fichier JSON, pas de constantes hardcodées) :

-- Auto-créées au boot via @mostajs/api-keys
CREATE TABLE api_key_scopes (
  id, name, label, description, icon, cardinality,
  valuesSource ('static' | 'dynamic'), dynamicSourceRef
);

CREATE TABLE api_key_scope_values (
  id, scopeName, value, label, sortOrder, metadata
);

Chaque module enregistre ses scopes au boot :

import { registerScope } from '@mostajs/api-keys/server'

// Dans mosta-net/src/server.ts (boot)
await registerScope(dialect, {
  name: 'transports', label: 'Network transports',
  cardinality: 'low', valuesSource: 'static',
  staticValues: [
    {value:'rest', sortOrder:1},
    {value:'graphql', sortOrder:2},
    // … 9 more
  ],
})

await registerScope(dialect, {
  name: 'projects', label: 'Projects',
  cardinality: 'high', valuesSource: 'dynamic',
  dynamicSourceRef: 'Project.slug',  // queried at runtime
})

L'admin UI charge le catalogue dynamique via GET /api/api-keys/scopes et rend une matrice (composant React ApiKeyScopeMatrix dans @mostajs/api-keys/components/).


Multi-project support

Via @mostajs/mproject — N bases de données isolées sur le même serveur.

Routes path-prefix

# Default project (DB_DIALECT + SGBD_URI au boot)
curl http://localhost:4488/api/v1/User

# Project nommé 'analytics'
curl http://localhost:4488/api/v1/analytics/events

# Ou via header (équivalent)
curl http://localhost:4488/api/v1/events -H "X-Project: analytics"

Ajouter un projet à l'exécution

curl -X POST http://localhost:4488/api/projects \
  -H "X-API-Key: <admin-key>" \
  -d '{
    "name": "analytics",
    "dialect": "mongodb",
    "uri": "mongodb://localhost:27017/analytics",
    "schemas": [{"name":"Event", "fields":{"type":"string","ts":"date"}}]
  }'

Persisté dans projects-tree.json (chemin : MOSTA_PROJECTS env var).


Quick start (self-host)

Install

npm install @mostajs/net @mostajs/orm @mostajs/mproject \
            @mostajs/rbac @mostajs/auth @mostajs/api-keys \
            @mostajs/config better-sqlite3

Run with SQLite (zero infra)

DB_DIALECT=sqlite \
SGBD_URI=./data/octonet.db \
DB_SCHEMA_STRATEGY=update \
OCTONET_ADMIN_EMAIL=admin@example.com \
OCTONET_ADMIN_PASSWORD=ChangeMe123! \
npx mostajs-net serve

Console output :

Loaded 3 schemas from schemas.json
[DAL:SQLite] INIT_SCHEMA strategy=update {"entities":["User","Product","Order"]}
✓ Apikey scopes registered (projects, operations, transports)
✓ RBAC ready — admin=admin@example.com trial=…  public=…
⚠ public demo apikey emitted ONCE → sk_live_xxxxxxxx…  (save it!)
✓ Sanitizer middleware on rest, graphql, ws, sse, trpc, mcp, odata, jsonrpc
✓ ApiKey middleware on rest, graphql, ws, sse, trpc, mcp, odata, jsonrpc
✓ Protected ormHandler ready (sanitizer + apikey global wrapper)
✓ T1 sandbox endpoint /try ready (rate-limited 10/h/IP, TTL 7d)

  @mostajs/net  v2.6.x
  ─────────────────────────────────────────────────
  Dialect:    sqlite (./data/octonet.db)
  Entities:   User, Product, Order  (3)
  Port:       4488
  Strategy:   update
  Transports: rest, graphql, ws, sse, trpc, mcp, odata, jsonrpc  (8)
  Ready.  3 entities × 8 transports = 24 endpoints

Test

# Sans apikey → 401
curl http://localhost:4488/api/v1/User
# {"error":{"code":"UNAUTHORIZED","message":"API key required…"}}

# Avec apikey publique (depuis les logs ↑)
curl http://localhost:4488/api/v1/User -H "X-API-Key: sk_live_…"
# {"status":"ok","data":[{"id":"…","email":"admin@example.com",…}]}  ← password absent

Configuration (.env)

Database (mandatory)

DB_DIALECT=postgres                                # any of 13 dialects
SGBD_URI=postgresql://user:pass@localhost:5432/db
DB_SCHEMA_STRATEGY=update                          # 'update' (recommandé prod) | 'create' (drop+recreate, dev only)
DB_SHOW_SQL=false

System dialect (recommandé prod, optionnel) — v2.7.5+

Sépare la base système (apikeys, RBAC users, audit, plans, payments, project-life metadata) du dialect métier mutable. Sans ces variables, alias transparent vers le singleton métier (rétro-compat mono-base).

MOSTA_SYSTEM_DIALECT=postgres                       # any of 13 dialects
MOSTA_SYSTEM_URI=postgresql://user:pass@localhost:5432/octonet_system

Voir section System dialect ci-dessous pour la motivation et les how-to.

Server

MOSTA_NET_PORT=4488
MOSTA_PROJECTS=./projects-tree.json                # path to multi-project config
SCHEMAS_PATH=./schemas                              # directory scanning fallback

Transports (toggle each)

MOSTA_NET_REST_ENABLED=true
MOSTA_NET_GRAPHQL_ENABLED=true
MOSTA_NET_WS_ENABLED=true
MOSTA_NET_SSE_ENABLED=true
MOSTA_NET_JSONRPC_ENABLED=true
MOSTA_NET_MCP_ENABLED=true
MOSTA_NET_TRPC_ENABLED=true
MOSTA_NET_ODATA_ENABLED=true
MOSTA_NET_GRPC_ENABLED=false
MOSTA_NET_NATS_ENABLED=false
MOSTA_NET_ARROW_ENABLED=false

MOSTA_NET_CORS_ORIGIN=*
MOSTA_RATE_LIMIT_CLIENT=1000

RBAC bootstrap (lus via @mostajs/config — supports profile cascade MOSTA_ENV=DEV)

OCTONET_ADMIN_EMAIL=admin@example.com               # 1er boot only ; ignored after
OCTONET_ADMIN_PASSWORD=secret-12345
OCTONET_ADMIN_FIRSTNAME=Admin                       # optional
OCTONET_ADMIN_LASTNAME=Octonet                      # optional
OCTONET_BOOTSTRAP_VERBOSE=false                     # detailed bootstrap logs
OCTONET_OPEN_MODE=false                             # dev only — passe sans apikey

T1 sandbox (optional)

OCTONET_TRIAL_DATA_DIR=./data/trials                # SQLite files per alias

Cloud middleware (optional, futur — partage DB méta avec Octocloud)

OCTONET_META_URI=postgresql://user:pass@localhost:5432/octonet_meta   # shared meta DB
PORTAL_DB_URI=postgresql://user:pass@localhost:5432/octonet_cloud     # legacy octocloud DB

API endpoints reference

Public

Endpoint Méthode Auth Description
/health GET open server status + transports + entities
/try GET open T1 sandbox provisioning HTML form
/try POST open (rate-limit 10/h/IP) crée sandbox + retourne apikey one-shot
/api/v1/health GET open health under api-versioned path

Project / entity CRUD (requires apikey)

Endpoint Méthode Op Description
/api/v1/{Entity} GET findAll list entities of default project
/api/v1/{Entity}/:id GET findById get one
/api/v1/{Entity}/count GET count count
/api/v1/{Entity}/one GET findOne first match
/api/v1/{Entity}/search GET search text search
/api/v1/{Entity} POST create insert
/api/v1/{Entity}/:id PUT update replace
/api/v1/{Entity}/:id DELETE delete remove
/api/v1/{Entity}/:id/addToSet POST addToSet add to array
/api/v1/{Entity}/:id/pull POST pull remove from array
/api/v1/{Entity}/:id/increment POST increment numeric add
/api/v1/{Entity}/upsert POST upsert insert or update
/api/v1/{Entity}/aggregate POST aggregate pipeline
/api/v1/{Entity}/updateMany POST updateMany bulk update
/api/v1/{Entity}/deleteMany POST deleteMany bulk delete
/api/v1/{project}/{Entity}/… * * same routes scoped to a project

Schema management (requires admin apikey)

Endpoint Méthode Description
/api/upload-schemas-json POST push schemas at runtime (triggers reload)
/api/apply-schema POST apply schema diff
/api/compare-schema POST dry-run schema diff
/api/schemas-config GET current registered schemas

Auth & API keys (requires admin)

Endpoint Méthode Description
/api/api-keys GET / POST list / issue keys
/api/api-keys/:id PUT / DELETE update / revoke
/api/api-keys/scopes GET list registered scopes + values (for admin matrix UI)
/api/api-keys/scopes POST register a new scope (admin)
/api/api-keys/scopes/:name/values PUT / DELETE manage scope values

Transports natifs

Endpoint Description
/graphql GraphQL endpoint + GraphiQL IDE
/ws WebSocket connection (entity events)
/events SSE stream
/rpc JSON-RPC 2.0
/mcp Model Context Protocol (Claude/Smithery/etc.)
/trpc/{Entity}.{op} tRPC procedure
/odata/{Collection} · /odata/$metadata OData v4

Multi-project management

Endpoint Méthode Description
/api/projects GET list projects
/api/projects POST add project
/api/projects/:name PUT / DELETE edit / remove

Observability

Endpoint Description
/api/performance live metrics (req/s, p50, p99)
/api/config-tree configuration tree (interactive)
/api/live-log streaming log feed

Architecture

                                 ┌──────────────────────────────────┐
                                 │       Fastify (port 4488)        │
                                 └────────────────┬─────────────────┘
                                                  │
       ┌─────────────────────────────────────────┴──────────────────────────────┐
       │                                                                        │
       ▼                                                                        ▼
┌────────────────────────┐                                       ┌────────────────────────┐
│  Transports (11)       │                                       │  Direct Fastify routes │
│  - REST                │                                       │  - /try                │
│  - GraphQL (mercurius) │                                       │  - /api/projects/*     │
│  - WS                  │                                       │  - /api/api-keys/*     │
│  - SSE                 │                                       │  - /health, /metrics   │
│  - JSON-RPC            │                                       └────────────────────────┘
│  - MCP (SDK Anthropic) │
│  - gRPC                │              ┌─────────────────────────────────────┐
│  - tRPC                │              │  Auth pipeline                      │
│  - OData               │              │  ┌──────────────┐                   │
│  - NATS                │              │  │ extractAuth  │ ←  HTTP request   │
│  - Arrow               ├──────────────┤  │ Context      │                   │
└────────────────────────┘              │  └──────┬───────┘                   │
                                        │         ▼                           │
                                        │  ┌──────────────┐                   │
                                        │  │ checkRequest │ ←  scope-checks   │
                                        │  │ (mosta-auth) │                   │
                                        │  └──────┬───────┘                   │
                                        │         ▼                           │
                                        │  ┌──────────────┐                   │
                                        │  │ checkApiKey  │ ←  resolve+verify │
                                        │  │ (api-keys)   │                   │
                                        │  └──────┬───────┘                   │
                                        └─────────┼───────────────────────────┘
                                                  ▼
                                        ┌────────────────────────┐
                                        │ Sanitizer middleware   │
                                        │ strips password/hash/  │
                                        │ tokens before response │
                                        └────────┬───────────────┘
                                                 ▼
                                        ┌────────────────────────┐
                                        │ ormHandler             │
                                        │ → ProjectManager.resolve│
                                        │ → EntityService.execute │
                                        └────────┬───────────────┘
                                                 ▼
                                        ┌────────────────────────┐
                                        │ @mostajs/orm dialect   │
                                        │ (13 SGBD)              │
                                        └────────────────────────┘

Modules de l'écosystème consommés

Module Rôle dans Octonet
@mostajs/orm persistance (13 dialects)
@mostajs/mproject gestion multi-projet (default + sandbox + abonnés)
@mostajs/rbac identité (User, Role, Permission, Account) + AccountSchema + OCTONET_RBAC_SEED
@mostajs/auth checkRequest orchestrateur framework-agnostic, hashPassword
@mostajs/api-keys apikey CRUD + scopes (Scope/ScopeValue schemas) + checkApiKey + ApiKeyScopeMatrix admin UI
@mostajs/config env var helper avec cascade MOSTA_ENV
@mostajs/cloud-middleware quota / abonnement (optionnel — actif si Octocloud connecté)
@mostajs/replicator CQRS multi-replica (optionnel)
@mostajs/project-life persistence schemas Project (optionnel — si stockage SGBD vs JSON)

System dialect (séparé du singleton métier)

Disponible depuis v2.7.5 — résout le bug « apikeys introuvables après /api/change-dialect » observé en prod.

Pourquoi deux dialects distincts ?

Octonet expose des routes admin qui mutent la connexion DB au runtime (/api/change-dialect, /api/reload-config, /api/reconnect) — légitime côté métier (les entités userland du projet courant peuvent migrer postgres → sqlite → mongodb selon les besoins de l'admin).

Mais les modules système (apikeys, RBAC users, audit, plans de souscription, payments, project-life metadata) doivent vivre dans une base stable qui ne suit pas ces mutations. Sinon : un changement de dialect métier rend les apikeys introuvables, l'admin se trouve verrouillé hors de l'IHM.

Rôle Variable env Mutable au runtime ? Lu par
Métier (entités userland) DB_DIALECT + SGBD_URI Oui (IHM admin) EntityService, transports, routes data
Système (infra Octonet) MOSTA_SYSTEM_DIALECT + MOSTA_SYSTEM_URI Non (stable) RBAC, apikey-middleware, account-scope, auth guards, sandbox /try

Configuration recommandée (prod multi-base)

# Métier — peut bouger via /api/change-dialect
DB_DIALECT=postgres
SGBD_URI=postgresql://hmd:***@127.0.0.1:5432/octonet_business

# Système — stable, jamais touché par les routes admin
MOSTA_SYSTEM_DIALECT=postgres
MOSTA_SYSTEM_URI=postgresql://hmd:***@127.0.0.1:5432/octonet_system

Configuration mono-base (dev / déploiement simple)

Laisser MOSTA_SYSTEM_* vides → alias automatique vers le singleton métier (rétro-compat 100 %, comportement identique au pré-v2.7.5).

How-to

1. Migrer un déploiement existant vers le mode multi-base

# 1. Sauvegarder la base actuelle
pg_dump octonet_business > backup-pre-split.sql

# 2. Créer la base système (vide — RBAC seed re-exécute au boot)
createdb octonet_system

# 3. Ajouter MOSTA_SYSTEM_* dans .env
echo 'MOSTA_SYSTEM_DIALECT=postgres' >> .env.local
echo 'MOSTA_SYSTEM_URI=postgresql://hmd:***@127.0.0.1:5432/octonet_system' >> .env.local

# 4. Restart Octonet — le bootstrap RBAC + scopes seed sur octonet_system
pm2 restart octonet-mcp

# 5. Restaurer les apikeys et RBAC users de l'ancienne base
#    (à scripter selon convention métier — typiquement export/import des
#    tables api_keys, users, roles, permissions, scopes, scope_values)

2. Vérifier que le système est bien isolé du métier

# Avant /api/change-dialect : apikey doit fonctionner
curl -H "X-API-Key: sk_live_…" http://localhost:4488/api/auth/verify
# → 200 OK

# Bouger le dialect métier vers SQLite
curl -X POST -H "Content-Type: application/json" \
  -d '{"dialect":"sqlite","uri":":memory:","connect":true}' \
  http://localhost:4488/api/change-dialect
# → "Dialecte changé et connecté : sqlite"

# Re-tester l'apikey — DOIT toujours fonctionner (système intact)
curl -H "X-API-Key: sk_live_…" http://localhost:4488/api/auth/verify
# → 200 OK (bug résolu en v2.7.5)

Avant v2.7.5, le 2ᵉ curl retournait 503 metadata DB unavailable ou 401 PostgreSQL not connected. Call connect() first.

3. Lire le system dialect depuis du code applicatif

import { getSystemDialect } from '@mostajs/data-plug'

const sysDialect = await getSystemDialect()
// Utiliser comme un dialect normal pour requêter les tables système
// (apikeys, users, roles, audit_log, plans, etc.)

4. Inspecter le bootstrap au démarrage

Au boot d'octonet-mcp, deux logs apparaissent :

✓ DB connectée: postgres                            ← métier
✓ System dialect: postgres (octonet_system)         ← système (si MOSTA_SYSTEM_* défini)
✓ RBAC ready — admin=… trial=… public=…             ← seed côté système
✓ ApiKey middleware on REST/SSE/GraphQL/...         ← câblé sur système

Sites système dans src/server.ts (pour référence)

18 callers tirent désormais leur dialect via getSystemDialect() au lieu du singleton métier :

Bloc Sites Caller
RBAC bootstrap 1 bootstrapRbac(systemDialect, …)
Scopes register 4 registerScope(systemDialect, …) × 3 + systemDialect.initSchema(scopeTables)
Middlewares globaux 2 createApiKeyMiddleware(() => systemDialect, …) + createAccountScopeMiddleware(() => systemDialect)
Middlewares per-transport 2 idem appliqués sur chaque transport.use(…)
Auth guards transports 6 authGuard(systemDialect, …) × 6 (SSE, GraphQL, JSON-RPC, gRPC, tRPC, OData)
Custom /api/auth/verify 2 checkApiKey(systemDialect, …) + new UserRepository(systemDialect)
Sandbox /try 2 registerTryRoutes({ dialect: systemDialect, … }) + startTrialCleanupJob({ dialect: systemDialect, … })

Le dialect métier (dialect) reste utilisé légitimement pour :

  • Bootstrap initial du singleton métier (L98) + pm.setDefault('default', dialect, …) (L126, L240)
  • Routes admin /api/reconnect, /api/change-dialect, /api/reload-config, /api/test-connection, /api/truncate-tables, /api/drop-tables — toutes opérations explicitement métier
  • EntityService qui sert les entités userland (opérations CRUD via les transports protégés)

Étapes du chantier (historique)

Étape Repo Livré
1 @mostajs/data-plug v1.2.2-1.2.4 (API getSystemDialect + façade ORM) npm
2 @mostajs/net v2.7.5 (bootstrapSystemDialect au démarrage) git
3 @mostajs/api-keys 0.2.3, @mostajs/payment 0.4.1, @mostajs/project-life 0.1.3, @mostajs/subscriptions-plan 0.3.5 (WeakMap repos + façade) npm
4 @mostajs/net v2.7.5 (consumers basculent sur getSystemDialect() — 18 sites) git
5 Tests d'intégration scénario /api/change-dialect postgres → sqlite
6 Déploiement amia + smoke test MOSTA_SYSTEM_URI

Boot sequence

  1. Load .env via @mostajs/config (profile cascade MOSTA_ENV)
  2. Connect main dialect (SGBD_URI)
  3. Load schemas (priority : getAllSchemas() registry → schemas.jsonSCHEMAS_PATH directory scan)
  4. dialect.initSchema(schemas) with strategy update/create
  5. Bootstrap RBAC (octonet-rbac-bootstrap.ts) :
    • register UserSchema/RoleSchema/PermissionSchema/PermissionCategorySchema/AccountSchema/ApiKeySchema
    • seedRBAC(OCTONET_RBAC_SEED) — 6 categories, 25 permissions, 4 roles (admin/subscriber/trial/public)
    • createAdmin() from OCTONET_ADMIN_EMAIL/PASSWORD
    • create Account trial-playground (type='trial')
    • create User public-demo (role=public)
    • create Account public-system (type='system')
    • generate ApiKey public-default scoped to default project (read-only, REST + MCP) — emitted ONCE in clear
  6. Register canonical scopes : projects (dynamic, Project.slug), operations (static : read/write/admin), transports (static : 11 values)
  7. Push default project to projects-tree.json (ownerId=admin, visibility=public)
  8. Load additional projects from projects-tree.json
  9. Wrap ormHandler with composeMiddleware([sanitizer, apikey], ormHandler)protectedOrmHandler
  10. Start each enabled transport :
    • transport.use(loggingMiddleware)
    • transport.use(sanitizerMiddleware)
    • transport.use(apiKeyMiddleware)
    • transport.setHandler(ormHandler)
    • transport.start(config)
  11. Register Fastify routes :
    • registerDynamicRestRoutes(app, protectedOrmHandler, pm)/api/v1/...
    • registerProjectRoutes(app, pm, protectedOrmHandler)/:project/*
    • registerTryRoutes(app, {dialect, pm})/try POST
    • registerTryPage(app)/try GET (HTML)
    • startTrialCleanupJob({dialect, pm}) — cron horaire TTL 7j
  12. Listen on port (MOSTA_NET_PORT)

14 polyglot NetClients

Octonet est consommable depuis 14 runtimes natifs, avec la même apikey, la même URL :

Runtime Package Registre
Node/TS @mostajs/net/client (built-in) npm
Java com.mostajs:mostajs-net-client Maven Central
Java + Spring Boot com.mostajs:mostajs-net-client-spring-boot-starter Maven Central
.NET MostaJs.Net.Client NuGet
Python mostajs-net-client PyPI
Go github.com/apolocine/mosta-net-client-go pkg.go.dev
Swift mosta-net-client-swift Swift Package Index
Kotlin io.github.apolocine:mostajs-net-client-kt Maven Central
Dart / Flutter mostajs_net_client pub.dev
Rust mostajs-net-client crates.io
PHP mostajs/net-client Packagist
Ruby mostajs-net-client RubyGems
Elixir mostajs_net_client Hex.pm
Lua mostajs-net-client LuaRocks
Delphi MostaJsNetClient GetIt
Unity (C#) com.mostajs.net-client OpenUPM

Monorepo : github.com/apolocine/mosta-net-clients

Chaque NetClient suit le même contrat d'API :

NetClient.create()
  .url(...)
  .apiKey(...)
  .build()
  .findAll(entity, filter, options)
  .findById(entity, id)
  .create(entity, data)
  .update(entity, id, data)
  .delete(entity, id)
  .uploadSchemasJson(schemas)
  .health()

Production deploy (amia.fr)

Le déploiement de référence tourne sur octonet.amia.fr + mcp.amia.fr (alias compat) en backend PostgreSQL :

URL Rôle
https://octonet.amia.fr Octonet server (REST, MCP, GraphQL, WS, SSE, tRPC, OData, JSON-RPC)
https://mcp.amia.fr alias DNS de compatibilité — historique mcp.so
https://octocloud.amia.fr Octocloud — portail SaaS Next.js (subscriptions, admin UI)

Le kit de déploiement complet est dans Entreprise/octonet-mcp/ :

  • apache/mcp.amia.fr.conf — vhost Apache2 (proxy SSE, WS upgrade, certs Let's Encrypt SAN)
  • ecosystem.config.cjs — config PM2
  • deploy.sh / install.sh / update.sh — scripts d'orchestration
  • tests/smoke-all.sh — suite de smoke tests (DNS, TLS, /health, SSE MCP, REST, /try, NetClient Java)

CLI

npx mostajs-net serve                      # Start server
npx mostajs-net mcp                        # MCP-only mode
npx mostajs-net generate-apikey <label>    # Émettre une apikey via CLI
npx mostajs-net hash-password <password>   # Hash bcrypt
npx mostajs-net info                       # JSON config dump

npx octonet-mcp --dialect=X --uri=Y        # Standalone MCP server (process séparé)

License

AGPL-3.0-or-later — usage libre tant que le code dérivé reste open-source. Licence commerciale disponible : drmdh@msn.com. Pricing par projet, pas par seat.

— (c) 2026 Dr Hamid MADANI <drmdh@msn.com>

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured