# SentientOne > SentientOne is an AI agent platform for teams and developers. Build, deploy, and manage custom AI agents with one API. Connect your tools and data via MCP (Model Context Protocol). Switch LLM providers and models from a dashboard — your application code never changes. SentientOne handles the operational layer of shipping AI in production: prompt caching, automatic retries, rate-limit and token management, cost optimization, and provider/model switching. Customers ship AI features in hours instead of weeks, without an in-house AI engineering team. Built by Infonex Pty Ltd (ABN 93 632 080 427) in Australia. Plans start at USD $19/month with a 14-day free trial. BYOK (bring your own LLM API keys) on every plan. ## Core concepts - **AI Agents** — Custom agents with their own system prompt, model, temperature, and tools. Isolated per account. Soft-delete and restore at any time. - **Knowledge** — Per-agent knowledge base with documents (PDFs), FAQs, and web crawling. Replies cite sources. - **MCP Integration** — Expose internal REST/gRPC APIs via the Model Context Protocol. Agents auto-discover available tools. - **Chatbot Widget** — One line of code to embed a branded chatbot on any website (WordPress, Shopify, custom HTML). - **AI Workspace** — Private chat portal for teams. Grounded in your data, no leaks to public AI. - **API Integration** — Every agent has a REST endpoint. Authenticate with `sk-so-…` keys. Chat and streaming endpoints. - **API Tracing** — End-to-end execution flow per request: auth, retrieval, tool calls, LLM, latency, tokens, cost. - **Chat & Testing** — Test agents directly from the dashboard, real-time streaming, full history. ## Pages - [Home](https://sentientone.ai/): Product overview, hero, capabilities, pricing. - [AI Agents](https://sentientone.ai/ai-agents): Create custom AI agents with prompts, models, tools. - [Knowledge](https://sentientone.ai/knowledge): Per-agent document, FAQ, and web-crawl knowledge base. - [MCP Integration](https://sentientone.ai/mcp-integration): Connect internal APIs to agents via Model Context Protocol. - [Chatbot Widget](https://sentientone.ai/chatbot): Embed AI chatbots on any website with one snippet. - [AI Workspace](https://sentientone.ai/workspace): Private team chat portal trained on your data. - [API Integration](https://sentientone.ai/api-integration): Public REST API for every agent. Chat and streaming. - [API Tracing](https://sentientone.ai/api-tracing): Per-request observability — latency, tokens, cost. - [Chat & Testing](https://sentientone.ai/chat-testing): Test and iterate on agents from the dashboard. - [Use Cases](https://sentientone.ai/use-cases): Production patterns — order status, support automation, BI, product discovery, recommendations. - [Blog](https://sentientone.ai/blogs): Articles on AI agents, agentic AI, MCP, and production AI rollout. ## Pricing - **Starter** — USD $19/month. 1 agent, 10k API requests/month, chatbot widget, 100 knowledge credits, 30 RPM, 14-day log retention, community support. - **Pro** — USD $49/month. 5 agents, 50k API requests/month, 300 knowledge credits, 50 RPM, 90-day log retention, full observability, AI Workspace, streaming endpoint, email support (48h SLA), team members at $9/each. - **Enterprise** — Custom pricing. Unlimited agents, MCP servers, knowledge, custom rate limits, self-hosted option, dedicated SLA, organisational onboarding. All plans include BYOK (bring your own LLM API keys), the chat interface, and a 14-day free trial. Prices in USD. Cancel anytime. ## Supported LLM providers GPT-4o (OpenAI), Claude (Anthropic), Gemini (Google), and any OpenAI-compatible model. Switch models from the dashboard — your application code never changes. ## Optional - [Documentation](https://app.sentientone.ai/docs): Developer-facing API and SDK docs. - [Application](https://app.sentientone.ai): SentientOne dashboard for managing agents. - [Built by Infonex](https://infonex.com.au): Company behind SentientOne. - [Privacy Policy](https://sentientone.ai/privacy) - [Terms of Use](https://sentientone.ai/terms)