[grade=A] Add DashCaddy blog and LLM install map

This commit is contained in:
Krystie committed 2026-10-05 20:16:33 -07:00
1 parent 22c3409c75
commit 5c71f251ca
10 files changed
+476 -1

No files matched your search

@@ -0,0 +1,83 @@
Self-hosting has always carried a tax: every app you want to run means a new
container to configure, a reverse proxy to wire up, a certificate to issue, and a
dashboard to babysit. DashCaddy was built to erase that tax — and the clearest
example is the **AI Intent Router**, which lets you describe what you want in
plain English and have it deployed, secured, and running.
## From a sentence to a running service
Traditional self-hosting asks you to think like an operator: pick an image, map
the ports, mount the volumes, set the environment, then hand-write a reverse-proxy
block and chase down a TLS certificate. DashCaddy collapses that into intent.
Instead of memorizing flags, you tell DashCaddy what you're trying to accomplish.
The AI Intent Router turns that plain-English request into a structured deploy
intent matched to one of DashCaddy's **77 pre-configured application templates**;
the deploy flow then provisions the service with sane defaults — container,
networking, and TLS via DashCaddy's built-in CA and Caddy. The result is a
working, HTTPS-secured app instead of a half-finished `docker run` command.
## 77 one-click templates, not a blank canvas
A blank Docker host is powerful and unforgiving. DashCaddy ships **77 one-click
templates** spanning media servers, productivity tools, developer platforms,
networking and DNS utilities, and more. Each template encodes the knowledge that
usually lives in a dozen browser tabs: a working default port and a health check,
wired and ready to run.
Because the templates are curated rather than improvised, "install Nextcloud" or
"set up a Git server" becomes a single action — deployable from the dashboard or
the REST API.
## Built for AI assistants: the MCP server
DashCaddy doesn't just use AI internally — it exposes itself *to* AI. The built-in
**MCP (Model Context Protocol) server** gives an AI assistant a set of named tools
— among them listing your services, checking service and system health, searching
the app catalog, and running container actions — over a standard protocol. Instead
of translating intent into API calls by hand, you can let an assistant do the
work. Self-hosting becomes something you can delegate.
## A real platform underneath: 400+ registered REST route handlers
The friendly front end sits on top of a genuine platform. DashCaddy registers
**400+ REST route handlers** covering deployment, lifecycle, health, security,
backups, and fleet operations — with a WebSocket channel for live updates.
Scripts and CI pipelines get a first-class path too. The AI layer is a
convenience, not a lock-in: the API is always there for the operators who want
it.
## Security that isn't an afterthought
Self-hosting is only worth it if it's safe to expose. DashCaddy includes a
**Security Center** — a dashboard for security events and the hosts it tracks —
so security visibility isn't a separate project you keep meaning to set up. TLS
is handled by the built-in CA and Caddy. And when you do want to let someone in,
**Secure Share Links** issue time-boxed public links (with an optional status
subscription) or single-use Tailscale invitations, so sharing a service doesn't
mean opening it to the world.
## Scales from one box to a fleet
What starts as a single-host homelab rarely stays that way. DashCaddy includes a
deploy bridge (its shipdeck integration) with live health re-probes and fleet
credentials for reaching other hosts, so growing beyond one machine doesn't mean
starting over with a new tool.
## Speaks your language — literally
DashCaddy's interface is localized into **31 languages**, including full
right-to-left support for Arabic. Self-hosting shouldn't be gated behind English
fluency, and the dashboard reflects that.
## Why this matters
The promise of self-hosting is control: your data, your infrastructure, your
rules. The cost has always been complexity. By turning intent into deployment,
curating 77 templates instead of handing you a blank canvas, and exposing 400+
registered REST route handlers plus an MCP server, DashCaddy keeps the control
while removing most of the complexity.
Ready to try it? Browse the [full feature set](/features), see the
[77-app catalog](/docs/catalog), or jump straight to the
[installation guide](/docs/installation).
+27
View File
@@ -0,0 +1,27 @@
// Blog post manifest — single source of truth for the index and the
// [slug] static-params generator. To publish, insert a new entry into the
// POSTS array (newest first) and add a matching markdown file at
// src/content/blog/<slug>.md. The build fails if a slug has no matching file.
export interface PostMeta {
slug: string;
title: string;
description: string;
date: string; // ISO yyyy-mm-dd
author: string;
tags: string[];
readingMinutes: number;
}
export const POSTS: PostMeta[] = [
{
slug: "deploy-self-hosted-apps-with-ai",
title: "Deploy Self-Hosted Apps From Plain English: DashCaddy's AI Intent Router",
description:
"How DashCaddy turns a plain-English request into a running, SSL-secured container — 77 one-click app templates, an MCP server for AI assistants, and 400+ registered REST route handlers.",
date: "2026-10-05",
author: "The DashCaddy Team",
tags: ["self-hosting", "ai", "docker", "homelab"],
readingMinutes: 6,
},
];