moorcheh-cookbooks

Moorcheh Cookbooks

Overview

This skill provides an index of implementation guides and foundational requirements for building Moorcheh-powered AI applications. Use the references to quickly scaffold full-stack applications with best practices for data management, semantic search, and RAG.

Moorcheh Account

If the user does not have an account yet, direct them to the cloud console to register and create a free account.

Create a Moorcheh account at console.moorcheh.ai.

Before Building Any Cookbook

Follow these shared guidelines before generating any cookbook app:

• Project Setup Contract
• Environment Requirements

Code Generation Rules

1. No Unicode emoji in output. Do not use ✅ ❌ 📁 ⏳ 🎉 or any emoji in print() or log messages. They crash on Windows cp1252 terminals. Use [OK], [ERROR], [WAIT] instead.
2. Use snake_case everywhere. Moorcheh JSON (REST and Python SDK) uses snake_case for answer/search payloads as of platform 1.5.10. Do not use legacy camelCase field names.

Then proceed to the specific cookbook reference below.

Cookbook Index

• Knowledge Base RAG: Build a document Q&A system that ingests documents into Moorcheh and generates AI-powered answers with source citations using RAG.
• Customer Support Bot: Build a customer support chatbot that answers questions from your FAQ and documentation using conversational RAG with chat history.
• Semantic Search App: Build a semantic search application with ITS scoring, metadata filtering, and relevance-labeled results.
• AI Q&A System: Build a question-answering system with structured output, custom prompts, and multi-namespace search.
• LLM Wiki- Self-maintaining personal knowledge base using Karpathy's LLM Wiki pattern extended with Moorcheh ITS search. See references/llm_wiki.md.
• Deep Ingest: Ingest large documents (>200K chars) or binary files (PDF, DOCX, XLSX) by uploading to a Moorcheh staging namespace. No local extraction needed- Moorcheh handles parsing, chunking, and indexing.

Integrations

• LangChain Integration: Use Moorcheh as a LangChain vector store for building chains and agents.

LLM Wiki (Karpathy Pattern + Moorcheh)

Build a self-maintaining personal knowledge base where an AI agent reads your documents once, builds a structured wiki of interlinked markdown pages, and uploads them to Moorcheh for persistent ITS-powered semantic search.

When to use this cookbook:

• You are accumulating knowledge across many sources over time (research, product, competitive intel)
• You want your AI agent to build and maintain a wiki automatically
• You need semantic search + metadata filtering across a large collection of notes
• You want knowledge to compound- every source and every query enriches the base

Key components:

• raw/- immutable source documents (agent reads only)
• wiki/- agent-generated markdown pages, interlinked and growing
• CLAUDE.md / AGENTS.md- the schema that tells the agent how to maintain the wiki
• Moorcheh namespace- persistent ITS-indexed copy of all wiki pages

Full implementation guide: references/llm_wiki.md Starter repo: https://github.com/moorcheh-ai/llm-wiki

Interface (Optional)

Use this when the user explicitly asks for a frontend for their Moorcheh backend:

• Frontend Interface: Build a Next.js frontend to interact with the Moorcheh backend, including chat UI and search interfaces.