> ## Documentation Index > Fetch the complete documentation index at: https://ona.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Getting started with agents > Set up agents with project context, integrations, and organization controls. Agents interface showing AI-assisted code editing with environment context Agents in Ona help you plan and apply code changes with full environment context and guardrails. Codex Agent is the recommended agent for new sessions and automations. ## Quick start 1. Start with [Codex Agent](/docs/ona/agents/codex). If your Enterprise organization still uses the deprecated Anthropic harness, review the [Ona Agent migration guidance](/docs/ona/agents/overview). 2. Add project context with [AGENTS.md](/docs/ona/agents-md). 3. Create reusable workflows with [Skills](/docs/ona/skills) and [Bash commands](/docs/ona/bash-commands). 4. Add optional integrations with [Model Context Protocol (MCP)](/docs/ona/mcp). 5. Run in [isolated environments](/docs/ona/runners/overview) or deploy the [Enterprise Runner](/docs/ona/runners/aws/enterprise-runner/overview). ## Security & controls * Enforce execution policies with the [command deny list](/docs/ona/command-deny-list) * Configure organization-wide [guardrails](/docs/ona/guardrails/overview) ## Next steps * [Codex Agent](/docs/ona/agents/codex): use the open-source Codex harness in Ona * [Ona Agent (deprecated)](/docs/ona/agents/overview): migrate workloads from the Anthropic harness * [Connect Codex to ChatGPT](/docs/ona/integrations/configure-codex): use your ChatGPT plan for Codex model requests * [LLM providers](/docs/ona/agents/llm-providers/overview): default model access and custom providers * [AGENTS.md](/docs/ona/agents-md): configure project context * [Skills](/docs/ona/skills) and [Bash commands](/docs/ona/bash-commands): automate workflows * [MCP](/docs/ona/mcp): integrate external tools