Working With AI Agents

Use coding agents to design, build, set up and review applications on Ovok, with the Ovok skill, the Ovok MCP server and AI-readable docs.

AI coding agents can help you explore Ovok, plan an integration, set up a project and implement application code. They work best when you give them the context of your application and ask them to verify platform behavior against the current Ovok documentation.

Runtime AI is a different topic

Building AI into your application at runtime, such as clinical summarization or agents acting on patient data, is a different topic; see Build with AI on Ovok for Ovok's documented $ai feature. This section is about using AI to develop on Ovok.

Three pieces

PieceWhat it gives the agentPage
Ovok skillGuidance: how to work on an Ovok app, what to verify, what to leave to a project admin. Not yet published for public install.Ovok skill
Ovok MCP serverAccess: named tools that read and set up one project, as the admin who connected it.Connect the MCP server · MCP tools
AI-readable docsContext: Markdown for every page, a docs index, a page manifest and a full bundle.Docs for AI agents

The docs work with any coding agent and need no Ovok access. The MCP server needs a project admin to approve it.

Quick start

  1. Connect the MCP server for the sandbox, then authenticate it with /mcp in Claude Code:

    claude mcp add --transport http ovok https://api.sandbox.ovok.com/mcp
  2. Ask for the work, with enough context. For example: “Read the project with the Ovok MCP tools, then read the Ovok docs for authentication and access policies, and plan a practitioner login for this app. Don't change settings yet.”

Then follow Prompting and review to plan, build and check the result.

Last updated on

On this page