navree

Ecosystem

Everything first‑party.Everything included.

Monitoring, deployment, an admin panel, queue supervision, search, billing, OAuth, AI and more are part of Navree. They are open source, they run on your servers, and they are designed, documented and released with the framework.

Manifestnavree 0.1 · MIT OR Apache-2.0
Some of what comes with Navree, where it lives in your app, and whether it is included
Package or toolWhat it isWhere it livesIncluded
Lighthouse/_lighthouseIncluded
Fleet/_fleetIncluded
Harbornavree deployIncluded
Helm/adminIncluded
Logbook/_logbookIncluded
Navigatornavree mcpIncluded
Workspacesnavree-workspaceIncluded
Searchnavree-searchIncluded
Billingnavree-billingIncluded
OAuth servernavree-oauthIncluded
AI SDKnavree-aiIncluded
Microservicesnavree-microIncluded
and 21 more, below
paid add-ons 0accounts 0
33first-party packages and tools
Documented here, built to the framework's conventions.
$0paid add-ons
No tiers, no seats, no hosted plan. MIT OR Apache-2.0.
1version number
Tested, documented and released together.
100%self-hosted
Dashboards run in your app; the data stays on your servers.

01 Run in production

See what production is doing, and ship to it without downtime.

Monitoring, queue supervision and deployment, built into the framework and hosted by you.

/_lighthouse

Lighthouse

Application monitoring, built in.

Exceptions, slow requests, slow queries, failed jobs and outgoing HTTP, grouped into issues and shown in a dashboard inside your app. Alerts go to email, Slack or a webhook, scripts read and resolve issues through its JSON API, and your coding agent through its MCP tools.

  • Issues grouped by fingerprint
  • Slow request and query traces
  • Alerts to email, Slack and webhooks
  • JSON API and MCP tools for agents
  • Dashboard served by your app
Read the Lighthouse docsnavree::LighthouseModule::from_config()
The Lighthouse dashboard of the example CRM: requests per minute, error rate, p95 latency, failed jobs and open issues; a 24-hour timeline of requests and exceptions; issues such as a slow contacts query on the deal board and a declined card when charging an invoice; and a live feed of new occurrences.
/_fleet

Fleet

Queue dashboard and worker supervisor.

Run supervised workers per queue that balance themselves under load. Watch throughput, wait times and runtimes, and retry or forget failed jobs from one dashboard.

  • Auto-balanced workers
  • Throughput and wait-time metrics
  • Retry or forget failed jobs
  • Pause and continue queues
  • Long-wait alerts
The Fleet dashboard of the example CRM: two supervisors running queue workers. When a newsletter puts 3,000 mails on the mail queue, Fleet moves processes to it, from one up to six, while the default and notifications queues keep what they need; below, mail jobs per minute spike at 09:00, and a list of failed jobs offers Retry and Forget.
navree deploy

Harbor

Servers and zero-downtime deployments.

Provision a server over SSH with TLS, PostgreSQL, Redis and systemd units for your web, worker and scheduler processes, then ship atomic releases that only go live after a health check.

  • Provision any Ubuntu server
  • Atomic, health-checked releases
  • One-command rollbacks
  • Encrypted environment secrets
  • Container images and push-to-deploy
Harbor deploying the example CRM with navree deploy: it builds, uploads, migrates, starts the new release in the green slot, waits for GET /health/ready to answer 200, switches Caddy from blue to green, drains blue and restarts the workers. Beside the terminal, the server app-1 shows its blue and green slots and the releases kept for rollbacks.

02 Build faster

Start with a product, not an empty directory.

Starter kits, an admin panel, a local debugger and local services, ready on day one.

/admin

Helm

An admin panel generated from your entities.

Describe a resource once and get list, detail and edit screens with filters, bulk actions and metrics. Every screen respects the access rules your API already enforces.

  • Resources from entities
  • Filters, actions and metrics
  • Uses your access policies
  • Custom fields and pages
Read the Helm docsnavree make:resource deals Deal
Helm, the admin panel of the example CRM, on the Deals resource: a sidebar with the CRM's resources, metric cards for the open pipeline, new deals and deals by stage; search and filters for stage and owner; a table of deals with amount, stage, company, owner and expected close date. Three deals are ticked and the bulk action Mark as won runs on them after a confirmation.
/_logbook

Logbook

A local debugging timeline.

Every request in development with the queries it ran, the events it fired, the jobs it queued, the mail it sent and the log lines it wrote, in one timeline you can filter and replay.

  • Requests, queries, jobs, mail and events
  • Filter by tag or user
  • Replay a request
  • Development only
Read the Logbook docsnavree::LogbookModule::from_config()
Logbook in development, open on the request POST /api/v1/deals: a list of recent requests and jobs on the left, one marked for duplicate queries; on the right the request's summary and its timeline, with the access check, each query, the deal.created event, the queued job and mail, the cache call, a log line and the 201 response, drawn against its 24 milliseconds.

03 Product capabilities

Search, billing, sign‑in and flags, ready when you need them.

The features every product grows into, designed to work with workspaces, access control and the queue.

05 AI

An MCP server for your coding agent, an AI SDK for your product.

For the agents that write your code and the AI features inside your product.

Start

Start with all of it.

The SaaS starter kit writes an app with sign-in, workspaces, billing, settings and an admin panel, wired to Navree. Lighthouse and Logbook come with every new app; every other tool is one module away.

$ cargo install --git https://github.com/NavreeHQ/navree navree-cli$ navree new acme --kit saas