AI Agents

Build Intelligent AI Agents Without Writing Code

Define what your agent knows, which tools it can use, and how it reasons. Pick from 19+ frontier LLM models, attach MCP servers for live data access, and deploy agents that stream answers in real time — no programming required.

19+ LLM Models — One Interface

Every agent gets its own model assignment. Use a lightweight model for simple FAQ bots and a frontier model for complex reasoning — optimizing cost and quality across your organization.

Claude Opus 4

Anthropic's most capable model for complex analysis, long-form content generation, and nuanced reasoning. Ideal for agents that need to synthesize large documents, write detailed reports, or handle multi-step logic chains where accuracy is paramount.

Claude Sonnet 4

The best balance of intelligence and speed in the Claude family. Sonnet handles most enterprise tasks — summarization, data extraction, customer support, code review — at a fraction of Opus cost with response times under two seconds.

Claude Haiku 3.5

Anthropic's fastest and most affordable model. Deploy Haiku for high-volume, low-latency use cases like chat widgets, classification tasks, and simple Q&A agents where sub-second response time matters more than deep reasoning.

GPT-5

OpenAI's flagship model offering state-of-the-art performance across coding, math, and creative writing. Excels at tool use and structured output generation. A strong choice for agents that need to produce JSON, call APIs, or write production-quality code.

GPT-4.1

Optimized for instruction-following and long-context tasks with a 1M-token context window. GPT-4.1 is well-suited for agents that process lengthy documents, codebases, or conversation histories where retaining every detail matters.

o3 & o4-mini

OpenAI's reasoning-optimized models that think step by step before answering. Use them for agents that tackle math problems, multi-constraint planning, scientific analysis, or any task where chain-of-thought reasoning dramatically improves accuracy.

Plus GPT-4o, GPT-4o-mini, o3-mini, and more. New models are added within days of release — your agents always have access to the latest capabilities. Learn how to choose the right model →

Configure Every Aspect of Your Agent

Each agent in Orckai is defined by four building blocks: a system prompt that sets its personality and instructions, a model selection that determines its intelligence, tools that give it real-world capabilities, and an optional knowledge base for domain-specific answers.

  • System Prompts — Write detailed instructions that shape how your agent behaves. Define its role, tone, boundaries, and output format. Prompts support variable interpolation with {{variable}} syntax for dynamic personalization.
  • Tool Attachment — Give your agent hands. Attach MCP tools for database queries, built-in tools for web search and calculations, or utility tools for sending emails and creating Jira tickets — all without writing integration code.
  • Knowledge Base (RAG) — Upload PDFs, Word documents, Excel spreadsheets, and 50+ other file types. Orckai chunks, embeds, and indexes them so your agent retrieves relevant context before every response, with inline source citations.
  • Per-Agent Model Selection — Assign different LLMs to different agents. Your customer support agent can use fast, affordable Haiku while your research agent uses powerful Opus — optimizing both cost and quality.
Create Your First Agent
Agent Definition
name: "Customer Support Agent"
model: claude-sonnet-4-20250514
system_prompt: |
You are a helpful support agent for Acme Corp.
Answer questions using the attached knowledge base.
Always cite your sources with [Source: filename].
If unsure, say so — never fabricate answers.
tools:
- mcp/acme-postgres (database queries)
- builtin/web-search
- utility/send-email
knowledge_base: "Product Documentation"

Two Execution Modes for Every Situation

Not every task requires the same level of reasoning. Orckai gives you two distinct runtime engines so you can match the execution mode to the complexity of the job — keeping things fast when they should be fast, and thorough when they need to be thorough.

  • Standard Mode (V2) — A streamlined runtime for straightforward tasks. The agent receives the user query, relevant context, and available tools, then produces a single coherent response. Ideal for FAQ bots, data lookups, document summarization, and customer support agents where speed matters.
  • Super Mode (SuperV2) — An enhanced runtime for complex multi-step reasoning. Super mode gives the agent expanded context management, the ability to chain multiple tool calls within a single turn, and deeper introspection on intermediate results. Use it for research agents, data analysis pipelines, or any task that requires planning, executing, and synthesizing across several steps.
  • Seamless Switching — Change the execution mode of any agent at any time with a single dropdown selection. No code changes, no redeployment. Test both modes on the same agent to find the right balance of speed and capability.
Execution Comparison
Standard (V2)
Query → Context + Tools → Single Response
Best for: FAQ, lookups, simple Q&A
Avg response: ~1-3 seconds
Super (SuperV2)
Query → Plan → Tool Call 1 → Reflect → Tool Call 2 → Synthesize → Response
Best for: Research, analysis, multi-step tasks
Avg response: ~5-15 seconds (depth varies)

Connect Your Agent to the Real World

An AI agent without tools is just a chatbot. Orckai lets you attach three categories of tools to any agent, turning it into an autonomous worker that can query databases, call APIs, search the web, send emails, and manipulate files — all governed by the permissions you define.

  • Custom MCP Servers — Generate MCP servers for PostgreSQL, MySQL, SQL Server, Oracle, MariaDB, and REST APIs. Deploy them as Docker containers with automatic networking. Your agent can run live database queries and API calls without exposing credentials to the model.
  • Built-in Tools — Pre-configured tools for common tasks: web search for real-time information, calculator for numerical operations, and date/time utilities. Available to any agent with a single toggle.
  • Utility Tools — Action-oriented tools for business automation: send emails via SMTP, create and update Jira tickets, read and write files, make HTTP requests, and parse structured data. Each tool is sandboxed with configurable timeouts and size limits.
  • Tool Discovery — MCP servers expose their available tools through a standard discovery protocol. When you attach an MCP server, Orckai automatically lists every tool the server provides. Your agent sees tool descriptions and parameter schemas and decides when and how to call them.
Explore MCP Servers
Attached Tools
MCP: acme-postgres
Tools: query_table, list_tables, describe_schema
Status: Running (Docker)
Built-in: Web Search
Tools: search_web, fetch_url
Status: Available
Utility: Send Email
Tools: send_email (SMTP)
Status: Configured
Utility: Jira Integration
Tools: create_issue, update_issue, search_issues
Status: Configured

Real-Time Streaming & Full Observability

Watch your agent think in real time. Every response streams token by token via Server-Sent Events. Every conversation is logged. Every token is counted and costed.

SSE Streaming Responses

Agent responses are delivered via Server-Sent Events (SSE) — the same protocol used by ChatGPT and Claude. Users see tokens appear in real time instead of waiting for a complete response. Streaming works across the web UI, embedded widgets, and the public API, giving every interface a responsive, conversational feel.

Persistent Conversation History

Every agent conversation is stored and resumable. Users can pick up where they left off, and agents maintain full context from prior turns. Conversation history is organization-scoped with row-level isolation, so multi-tenant deployments keep each team's data completely separate. Browse, search, and audit past conversations from the admin panel.

Token Usage & Cost Monitoring

Orckai tracks input tokens, output tokens, and estimated cost for every agent execution. View per-agent, per-user, and per-organization breakdowns. Set up alerts when usage exceeds thresholds. Detailed metrics help you right-size model selections — switch an over-provisioned agent from Opus to Sonnet and see the cost impact immediately.

Explore More Features

Build Your First AI Agent in Minutes

Pick a model, write a prompt, attach your tools, and deploy. No infrastructure to manage, no code to write, no vendor lock-in.