[Website] Broken links and cleaning (#9172)

GitOrigin-RevId: 70c0b97bf34aa08eac9de2f84da77afd581a1002

Author: olruas

docs/2.developers/4.user-guide/50.llm-xpack/20.llm-app-pathway.md

@@ -334,6 +334,6 @@ This real-time reactivity ensures that the RAG's responses are always based on t
 In this tutorial, you have learned to create a RAG pipeline from scratch with Pathway: you have implemented the pipeline step-by-step.
 The RAG is live by default, updating the index whenever the documentation changes.
 
-In addition to our [ready-to-run templates](/developers/templates?tab=ai-pipelines), Pathway provides pre-build wrapper to use [pre-selected prompts](/developers/api-docs/pathway-xpacks-llm/prompts), [servers](/developers/api-docs/pathway-xpacks-llm/servers), or even entire [RAG pipeline](/developers/api-docs/pathway-xpacks-llm/servers).
+In addition to our [ready-to-run templates](/developers/templates?tab=ai-pipelines), Pathway provides pre-build wrapper to use [pre-selected prompts](/developers/api-docs/pathway-xpacks-llm/prompts), [servers](/developers/api-docs/pathway-xpacks-llm/servers), or even entire [RAG pipeline](/developers/api-docs/pathway-xpacks-llm/question_answering).
 
 Instead of doing everything from scratch, take a look at the documentation, you might find what you are trying to do!

docs/2.developers/4.user-guide/50.llm-xpack/30.docs-indexing.md

@@ -24,12 +24,12 @@ Embedding transforms text into fixed-size vectors for indexing and retrieval.
 It is required only when using vector indices, such as approximate nearest neighbor (ANN) search. 
 Pathway provides several embedding models, including:
 
-- [`OpenAIEmbedder`](/developers/api-docs/pathway#pathway.xpacks.llm.embedders.OpenAIEmbedder)
-- [`LiteLLMEmbedder`](/developers/api-docs/pathway#pathway.xpacks.llm.embedders.LiteLLMEmbedder)
-- [`GeminiEmbedder`](/developers/api-docs/pathway#pathway.xpacks.llm.embedders.GeminiEmbedder)
-- [`SentenceTransformerEmbedder`](/developers/api-docs/pathway#pathway.xpacks.llm.embedders.SentenceTransformerEmbedder)
+- [`OpenAIEmbedder`](/developers/api-docs/pathway-xpacks-llm/embedders#pathway.xpacks.llm.embedders.OpenAIEmbedder)
+- [`LiteLLMEmbedder`](/developers/api-docs/pathway-xpacks-llm/embedders#pathway.xpacks.llm.embedders.LiteLLMEmbedder)
+- [`GeminiEmbedder`](/developers/api-docs/pathway-xpacks-llm/embedders#pathway.xpacks.llm.embedders.GeminiEmbedder)
+- [`SentenceTransformerEmbedder`](/developers/api-docs/pathway-xpacks-llm/embedders#pathway.xpacks.llm.embedders.SentenceTransformerEmbedder)
 
-More information can be find on the [`Embedders page`](/developers/user-guide/llm-xpack/embedders)
+You can find more information about embedders in the [`associated article`](/developers/user-guide/llm-xpack/embedders).
 
 ### Non-Vector Indexing
 
@@ -63,7 +63,7 @@ retriever_factory = BruteForceKnnFactory(
 
 ## How to query retriever?
 
-To interact with the index and retrieve relevant documents, we need to create a [`DocumentStore`](/developers/api-docs/pathway#pathway.xpacks.llm.document_store.DocumentStore).
+To interact with the index and retrieve relevant documents, we need to create a [`DocumentStore`](/developers/api-docs/pathway-xpacks-llm/document_store).
 This object will handle processing of documents (which include **parsing**, **post-processing** and **splitting**) and then building an **index** (retriever) out of them.
 The DocumentStore acts as the interface to query the index, allowing for document retrieval using the selected retriever.
 

docs/2.developers/4.user-guide/50.llm-xpack/40.pathway_mcp_server.md

@@ -10,7 +10,7 @@ keywords: ['MCP', 'MCP server', 'API', 'protocol']
 # Pathway MCP Server
 
 ## Introduction
-The Model Context Protocol (MCP) is designed to standardize the way applications interact with large language models (LLMs).
+The [Model Context Protocol](https://modelcontextprotocol.io/docs/getting-started/intro) (MCP) is designed to standardize the way applications interact with large language models (LLMs).
 It serves as a bridge, much like a universal connector, enabling seamless integration between AI models and various data sources and tools.
 This protocol facilitates the creation of sophisticated AI workflows and agents, enhancing the capabilities of LLMs by connecting them with real-world data and functionalities.