A RAG chatbot for websites, docs, and help centers that retrieves from your own content to inform answers. ChattyBox is designed to answer questions like "What is included in the Pro plan?" using your published pages and can provide source links — with no custom retrieval infrastructure to build or host.
RAG combines a language model with retrieved source content. Before generating a response, the chatbot searches your published pages for passages that match the question.
If your priority is answer trust, pair this retrieval workflow with a grounded AI chatbot designed to reduce invented answers through source constraints and fallback behavior.
When a visitor asks a question, ChattyBox searches indexed website, docs, help, and CMS content for relevant passages.
Those source passages are added as context so the model has something specific to answer from.
The model generates a concise answer and can include source links so visitors can verify the details.
A dependable RAG system is a chain of testable stages, not a single model call. The following generic reference architecture illustrates checks for content preparation, retrieval, generation, and citation; it is not a specification of ChattyBox capabilities.
Reviewed when retrieval behavior changes and at least every six months.
| Stage | Example operation | Control to verify |
|---|---|---|
Stage 1. Crawl | Example operation Discover allowed pages from URLs and sitemaps. | Control to verify Expected URLs are present; excluded and private paths are absent. |
Stage 2. Parse | Example operation Extract useful page text and retain source information for retrieval. | Control to verify Navigation noise is removed without dropping commands or version labels. |
Stage 3. Chunk | Example operation Split content into passages that retain their source page URL. | Control to verify Each passage is understandable alone and preserves prerequisites. |
Stage 4. Embed | Example operation Represent passages for semantic matching and store source metadata. | Control to verify Every vector maps to one current, permitted source passage. |
Stage 5. Retrieve | Example operation Rank a small set of passages against the visitor question. | Control to verify Recall@k tests put the gold passage in the context set. |
Stage 6. Prompt | Example operation Combine the question, passages, source metadata, and fallback rules. | Control to verify Review instructions for conflicts with grounding and fallback guidance; prompts alone cannot guarantee compliance. |
Stage 7. Answer | Example operation Instruct the model to use supported details or decline when evidence is insufficient. | Control to verify Check claims against the context; no usable context or a detected content gap can trigger a fallback. |
Stage 8. Cite | Example operation Return retrieved page URLs for supported answers; check that they support the response. | Control to verify Links resolve, match the version, and directly support the claim. |
Changed content must be fetched and stored before ChattyBox can use it. Vector retrieval depends on updated embeddings; in some cases, stored-page fallback can supply context without matching chunks. Refreshes and retrieval can still lag or fail, so test representative answers against current sources, especially for pricing, policy, and versioned docs.
Public pages can share one crawl scope. Private, tenant, employee, or account data needs authenticated ingestion and retrieval-time authorization before passages enter the prompt. Robots rules are not access control. Never index secrets, and test that unauthorized identities cannot retrieve titles, snippets, URLs, or answer details from restricted sources.
Use a versioned question set with known source passages, expected claims, and expected fallback cases. Run it after source, chunking, embedding, ranking, or prompt changes, then inspect failures by stage.
| Signal | How to measure it | Failure it exposes |
|---|---|---|
Signal Recall@k | How to measure it Share of answerable questions where a gold passage appears in the top k results. | Failure it exposes Missing crawl coverage, weak chunks, vocabulary mismatch, or stale embeddings. |
Signal Ranking quality | How to measure it Record reciprocal rank or nDCG for judged relevant passages. | Failure it exposes A relevant page exists but noise or an obsolete version outranks it. |
Signal Claim support | How to measure it Review every material claim against the passages supplied to the model. | Failure it exposes The answer combines assumptions with evidence or overstates the source. |
Signal Citation accuracy | How to measure it Verify that each citation resolves and directly supports its attached claim. | Failure it exposes The right answer links to a merely related page or the wrong version. |
Signal Fallback precision and recall | How to measure it Include unanswerable and conflicting-source questions; score correct declines and missed answers. | Failure it exposes The bot guesses without evidence or refuses despite adequate evidence. |
Signal Freshness lag | How to measure it Measure time from a published source change to a correct retrieved answer. | Failure it exposes Recrawl delay, duplicate stale chunks, or failed deletion propagation. |
| Method | Best fit | Important limit |
|---|---|---|
Method Keyword or site search | Best fit Users want ranked pages and will read the source themselves. | Important limit Does not synthesize an answer and can miss semantic matches. |
Method RAG | Best fit Knowledge changes, citations matter, and answers span a few passages. | Important limit Quality depends on crawl coverage, retrieval, permissions, and source quality. |
Method Fine-tuning | Best fit Teach stable style, format, classification, or task behavior. | Important limit Not a reliable replacement for current facts, citations, or source deletion. |
Method Long-context prompting | Best fit A bounded document set fits in context and every item is relevant. | Important limit Cost and noise grow with context; permissions and freshness still need controls. |
There is no universal best option. Test citation fidelity, fallback behavior, source freshness, conflicting content, escalation, private-context boundaries, analytics, integrations, and deployment effort against the same support questions.
A focused RAG chatbot can answer repeat questions from approved docs and help articles with citations and conservative fallback behavior.
Choose a broader support suite when you need authenticated account actions, ticket routing, SLAs, agent workspaces, or cross-channel operations.
Include missing, stale, conflicting, private, and high-impact questions. Record cited URLs, declines, handoffs, and reviewer notes instead of publishing an unverified score.
Plan the support and escalation boundary · Evaluate linked source evidence · Use the hallucination-reduction checklist
Suppose the indexed installation page says the widget script belongs immediately before the closing body tag. Retrieval should select that passage, the prompt should prohibit unsupported placement advice, and the answer should state the instruction with the installation page attached as evidence.
Continue with the guides to source citations, AI Chatbot for Docusaurus with Citations, developer documentation chatbots, and the MkDocs AI Chatbot for Source-Cited Docs.
Your website, docs, help center, and CMS pages are indexed so the chatbot can search the content you already publish.
Embeddings help match visitor questions to relevant source passages even when the wording is different.
ChattyBox searches for relevant source content to provide as context; if usable context is missing, it can return a fallback response.
The model is instructed to answer from retrieved context and avoid unsupported claims. These instructions do not guarantee a correct answer.
Supported answers can include retrieved source-page links for users to inspect. Fallback responses can have no citations.
When relevant content is missing, the bot can fall back instead of improvising an unsupported answer.
A generic chatbot answers from broad model memory and can sound confident about things it does not know. A RAG chatbot retrieves from your pages first and is instructed to use that content; source links can help users check the answer, but accuracy is not guaranteed.
When citations are the priority, pair retrieval with an AI chatbot with source citations that links supported answers to their source pages.
Connect your existing site content, test answer quality, and embed a source-cited chatbot when the responses look right.
Start Free - No Credit Card
