A retrieval chatbot over your own docs is one of the most one-shottable products there is: crawl the site, chunk and embed it, answer from the top matches with an LLM, drop in a widget. What you don't get for free is the boring operational layer, scheduled re-crawls, analytics, lead capture and human handoff, multi-source connectors, and a hosted widget that stays up. Buildable in a weekend, real gaps after that.
Der Moat
Ein Moat ist der Burggraben um ein Produkt: alles, was ein Nachbau nicht einfach mitkopieren kann, etwa Nutzer, Daten, Verträge oder Vertrauen. Je tiefer der Graben, desto eher lohnt sich das Abo trotz KI-Agent.Viele Schnittstellen brauchen OAuth, Pflege und zuverlässige Synchronisierung.
Zuverlässigkeit und sauberer Feinschliff sparen dauerhaft Arbeit.
Globaler Betrieb und hohe Verfügbarkeit sind schwer selbst zu tragen.
Der Prompt
Build me an AI support chatbot that trains on my own website and docs, to replace MagicChat, in an empty repo. Stack (no alternatives): Next.js 15 (App Router) + TypeScript, Postgres with the pgvector extension via Drizzle ORM, and Docker Compose so `docker compose up` runs Postgres and the app together. Use the OpenAI or Anthropic API for both embeddings and answers (keys in .env). Core loop: - `npm run ingest -- <sitemap-or-url>`: crawl the pages, strip to clean text, chunk (~800 tokens with overlap), embed each chunk, and store text + vector + source URL in Postgres. - A /api/chat route: embed the incoming question, pull the top-k chunks by cosine similarity, and ask the LLM to answer ONLY from that context, returning the source URLs it used. Stream the answer. - A single embeddable widget: one <script> tag mounts a floating chat bubble on any site, talking to /api/chat with CORS locked to configured origins. Details: - One config file: bot name, greeting, allowed origins, model, top-k. - Store everything locally in Postgres; `npm run reindex` re-crawls and replaces. - Secrets in .env, ship .env.example, never commit keys. - Handle empty, loading, and "I don't know from the docs" states honestly; never invent answers outside the retrieved context. - Out of scope: multi-channel (email/WhatsApp/Slack), team seats, an analytics dashboard, human handoff, scheduled auto-refresh (leave a documented cron hook), and any hosted control plane. - README: setup, the ingest command, embedding the widget, and where data lives.
Lies, was Entscheider lesen.
Jeden Sonntag das Wichtigste aus der KI-Welt. Kompakt, verständlich, praxisnah. Schließe dich über 4.500 Lesern an.
100 % kostenlos. Jederzeit mit einem Klick abmeldbar.
People pay so they never touch the plumbing: the crawler that re-indexes when docs change, the dashboard that shows what customers asked, the connectors to their help desk, and a widget that stays up without them running a server. The RAG is easy; keeping it fresh, measured and online is the recurring work.
Für den Eigenbedarf reicht der Eigenbau oben meistens.
| Plan | Monatlich | Jährlich (pro Monat) | Enthalten |
|---|---|---|---|
| Free | 0 € / workspace | 0 € | 1 chatbot; 100 messages/month; 50 training pages; website embed; community support |
| Lite | 25,01 € / workspace | n. a. | 1 chatbot; 7,500 messages/month; 500 pages; 1 team member; manual refresh |
| Starter | 50,88 € / workspace | n. a. | 1 chatbot; 15,000 messages/month; 1,000 pages; 1 team member; manual refresh |
| Growth | 111,25 € / workspace | n. a. | 2 chatbots; 33,000 messages/month; 10,000 pages; 4 team members; monthly auto-refresh; API access |
| Scale | 369,96 € / workspace | n. a. | 10 chatbots; 110,000 messages/month; 50,000 pages; 10 team members; weekly auto-refresh; daily auto-scan |
| Enterprise | n. a. / workspace | n. a. | Custom message volume, chatbot count, seats and compliance; HIPAA eligible with DPA/BAA on request |
Geprüft am 12.08.2026 · Quelle · aus USD umgerechnet, der EUR-Preis beim Anbieter kann abweichen
Preis, Nachbau und Ersparnis von MagicChat auf einen Blick.
MagicChat kostet im Plan „Starter" 50,88 € pro Monat (geprüft am 12.08.2026). Gratis-Version: 1 chatbot, 100 messages/month and 50 training pages
Jein, Wochenendprojekt: A retrieval chatbot over your own docs is one of the most one-shottable products there is: crawl the site, chunk and embed it, answer from the top matches with an LLM, drop in a widget. What you don't get for free is the boring operational layer, scheduled re-crawls, analytics, lead capture and human handoff, multi-source connectors, and a hosted widget that stays up. Buildable in a weekend, real gaps after that.
Beim Eigenbau verzichtest du auf: scheduled auto re-crawl and content refresh, analytics and conversation-history dashboards, lead capture and human handoff, multi-source connectors and integrations, hosted uptime for the widget, team seats and enterprise compliance (HIPAA/DPA/BAA).
Das Abo kostet 50,88 € pro Monat — aufs Jahr gerechnet sparst du rund 610,56 €.
Intercom
Kunden-Messenger mit Support-Inbox, Help Center und dem KI-Agenten Fin.
26 €/Monat
Tidio
Live-Chat und Ticket-System für Websites mit KI-Agent Lyro, der Standardfragen selbst beantwortet.
26 €/Monat
Crisp
Website-Chat mit geteilter Inbox, Wissensdatenbank und Chatbot für kleine Teams.
40 €/Monat
Der Kommentarbereich wird geladen, sobald du ihn erreichst.