# kapa.ai > kapa.ai turns technical documentation into production-grade, customer-facing AI assistants. Kapa indexes 50+ technical knowledge sources (documentation, source code, GitHub issues, support tickets, community forums, PDFs, and more) into a single unified knowledge base, then uses a retrieval-augmented generation (RAG) answer engine to respond to complex technical questions with high accuracy and inline source citations. Kapa is trusted by 200+ technical companies, including OpenAI, Docker, Grafana, Monday.com, Mapbox, CircleCI, and Nokia, and is SOC 2 Type II certified. Key facts for interpreting this site: - Kapa is model-agnostic. It continuously evaluates and uses the best-performing models from providers such as OpenAI, Anthropic, Cohere, and Voyage, and runs its own models where useful, rather than being tied to a single model. - Kapa is optimized for accuracy on technical content. Every answer is grounded in the customer's own sources, and Kapa responds with "I don't know" rather than hallucinating when its sources do not cover a question. - One knowledge base, many surfaces. Kapa deploys as a website "Ask AI" widget, Slack and Discord community bots, a support-ticket deflector, an internal employee assistant, a hosted MCP server, and an SDK/API for custom and in-product agents. - Two hosts are relevant. The marketing site is https://www.kapa.ai and the product documentation is https://docs.kapa.ai. Developer, integration, and configuration detail lives in the docs. - "kapa", "Kapa", and "kapa.ai" all refer to the same product. The legal entity is kapa.ai, Inc. ## Product - [kapa.ai home](https://www.kapa.ai/): Platform overview - connect your sources, the answer engine, and where you can deploy. - [Pricing](https://www.kapa.ai/pricing): How pricing works (usage-based, with a minimum annual commitment for commercial plans). - [Data security](https://www.kapa.ai/security): Security and privacy posture, including SOC 2 Type II, GDPR, SSO/RBAC, and PII redaction. - [Get a free trial](https://www.kapa.ai/get-trial): Start a 14-day free trial with an assistant trained on your content. - [Request a demo](https://www.kapa.ai/request-demo): Book a branded demo trained on your documentation. ## Documentation - [Documentation home](https://docs.kapa.ai/): Entry point for setup, integrations, and developer guides. - [Getting started](https://docs.kapa.ai/overview/getting-started): Step-by-step guide to deploying your first assistant. - [FAQ](https://docs.kapa.ai/overview/faq): What problem Kapa solves, which models it uses, language support, and multi-instance setup. - [Showcase](https://docs.kapa.ai/overview/showcase): Live, real-world deployments from companies like OpenAI, Nokia, and ClickHouse. - [Supported data sources](https://docs.kapa.ai/data-sources/supported-data-sources): The full list of 50+ connectors Kapa can index. - [Data sources overview](https://docs.kapa.ai/data-sources/overview): How ingestion, chunking, and automatic refresh work. ## Integrations and deployment - [All integrations](https://docs.kapa.ai/integrations): Every supported way to deploy Kapa. - [Website widget](https://docs.kapa.ai/integrations/website-widget): Embed an "Ask AI" chat widget on a docs or marketing site with a single script tag. - [Slack bot](https://docs.kapa.ai/integrations/slack-bot): Answer questions inside a Slack community or workspace. - [Discord bot](https://docs.kapa.ai/integrations/discord-bot): Answer questions inside a Discord community. - [Support form deflector](https://docs.kapa.ai/integrations/support-form-deflector): Suggest answers before a support ticket is submitted (resolves ~40% of tickets pre-submission). - [Internal technical assistant](https://docs.kapa.ai/integrations/internal-technical-assistant): An internal assistant for support, customer success, and solutions teams. ## Developers - [API overview](https://docs.kapa.ai/api/overview): HTTP API for custom integrations. - [SDK](https://docs.kapa.ai/dev/sdk/): React/JS SDK for building fully custom front-ends. - [Agent SDK](https://docs.kapa.ai/dev/agent): Build in-product agents that combine Kapa retrieval with your own custom tools. - [MCP server](https://docs.kapa.ai/integrations/mcp/overview): Kapa-hosted Model Context Protocol server for use in Claude, Cursor, ChatGPT, and other agents. Kapa's own documentation MCP server is public at https://kapa-docs.mcp.kapa.ai (Streamable HTTP), exposing the tool `search_kapa_knowledge_sources`. ## Analytics - [Analytics overview](https://docs.kapa.ai/analytics/overview): Understand what users ask and how the assistant performs. - [Coverage gaps](https://docs.kapa.ai/analytics/coverage-gaps): Find exactly where documentation is missing or insufficient. - [Common questions](https://docs.kapa.ai/analytics/common-questions): See the most frequent questions users ask. ## Customers and use cases - [Customer stories](https://www.kapa.ai/customer-stories): How companies like Netlify, Monday.com, Mapbox, and CircleCI use Kapa. - [In-app product AI assistant examples](https://www.kapa.ai/blog/in-app-product-ai-assistant-examples): How Mapbox, Monday.com, and CircleCI embed Kapa inside their products. - [How to build an AI agent that knows your product](https://www.kapa.ai/blog/how-to-build-an-ai-agent-that-actually-knows-your-product): Pairing native tools with retrieval via MCP. - [Open source program](https://docs.kapa.ai/kapa-for-open-source): Free Kapa for qualifying open source projects (up to 10,000 questions/month). ## Optional - [How Kapa prevents AI hallucination](https://www.kapa.ai/blog/ai-hallucination): The accuracy and hallucination-detection approach. - [Blog](https://www.kapa.ai/blog): Articles on RAG, technical AI assistants, and customer results. - [API catalog](https://www.kapa.ai/.well-known/api-catalog): Machine-readable API catalog (RFC 9727) linking the OpenAPI spec, API docs, and status page. - [MCP server card](https://www.kapa.ai/.well-known/mcp/server-card.json): Discovery metadata for Kapa's public documentation MCP server (SEP-2127). - [Terms of service](https://www.kapa.ai/content/terms-of-service) - [Privacy policy](https://www.kapa.ai/content/privacy-policy)