formio-uag — Claude Code MCP server: what it does, install
formio-uag connects Form.io forms to an AI agent through MCP. It turns form schemas into readable context and supports structured data collection.
claude mcp add formio-uag -e PROJECT=<PROJECT> -e PROJECT_KEY=<PROJECT_KEY> -e ADMIN_KEY=<ADMIN_KEY> -e UAG_LICENSE=<UAG_LICENSE> -e JWT_SECRET=<JWT_SECRET> -e PORTAL_SECRET=<PORTAL_SECRET> -e MONGO=<MONGO> -e BASE_URL=<BASE_URL> -- npx -y @formio/uagWhat formio-uag does
formio-uag exposes a Form.io form’s structure to an agent as readable Markdown. This supports two flows: turning conversational input into data that follows the form’s JSON schema, and running agentic workflows that use existing submissions together with extra context.
Form schemas can change the context used during processing without changing the application or retraining the agents. The server includes the agent_provide_data tool, while the repository examples cover a custom module, a conversation form, and an agentic workflow. The npx setup requires values for PROJECT, PROJECT_KEY, ADMIN_KEY, UAG_LICENSE, and JWT_SECRET; the first setup command also passes MONGO, BASE_URL, and PORTAL_SECRET.
What it helps with
- turn chat input into form submission data
- give an agent a readable Form.io schema
- analyse an existing submission with added context
- change agent context through a form JSON schema
Example prompts for the agent
- Read this Form.io form and extract the user’s answers into structured data.
- Use
agent_provide_datato analyse the existing submission against the supplied context. - Work out which fields this form needs and fill them from the conversation.
How to install formio-uag
Platform CLI
claude mcp add formio-uag -e PROJECT=<PROJECT> -e PROJECT_KEY=<PROJECT_KEY> -e ADMIN_KEY=<ADMIN_KEY> -e UAG_LICENSE=<UAG_LICENSE> -e JWT_SECRET=<JWT_SECRET> -e PORTAL_SECRET=<PORTAL_SECRET> -e MONGO=<MONGO> -e BASE_URL=<BASE_URL> -- npx -y @formio/uagReplace <…> with your own values. 10 more environment variables are in the server README
Platform CLI
codex mcp add formio-uag --env PROJECT=<PROJECT> --env PROJECT_KEY=<PROJECT_KEY> --env ADMIN_KEY=<ADMIN_KEY> --env UAG_LICENSE=<UAG_LICENSE> --env JWT_SECRET=<JWT_SECRET> -- npx -y @formio/uagReplace <…> with your own values. 10 more environment variables are in the server README
Config file
{
"mcpServers": {
"formio-uag": {
"command": "npx",
"args": [
"-y",
"@formio/uag"
],
"env": {
"PROJECT": "<PROJECT>",
"PROJECT_KEY": "<PROJECT_KEY>",
"ADMIN_KEY": "<ADMIN_KEY>",
"UAG_LICENSE": "<UAG_LICENSE>",
"JWT_SECRET": "<JWT_SECRET>"
}
}
}
}Replace <…> with your own values. 10 more environment variables are in the server README
Config file
{
"servers": {
"formio-uag": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@formio/uag"
],
"env": {
"PROJECT": "<PROJECT>",
"PROJECT_KEY": "<PROJECT_KEY>",
"ADMIN_KEY": "<ADMIN_KEY>",
"UAG_LICENSE": "<UAG_LICENSE>",
"JWT_SECRET": "<JWT_SECRET>"
}
}
}
}Replace <…> with your own values. 10 more environment variables are in the server README
Platform CLI
gemini mcp add -e PROJECT=<PROJECT> -e PROJECT_KEY=<PROJECT_KEY> -e ADMIN_KEY=<ADMIN_KEY> -e UAG_LICENSE=<UAG_LICENSE> -e JWT_SECRET=<JWT_SECRET> formio-uag npx -y @formio/uagReplace <…> with your own values. 10 more environment variables are in the server README
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about formio-uag
How do I install formio-uag in Claude Code?
Run `claude mcp add formio-uag -e PROJECT=<PROJECT> -e PROJECT_KEY=<PROJECT_KEY> -e ADMIN_KEY=<ADMIN_KEY> -e UAG_LICENSE=<UAG_LICENSE> -e JWT_SECRET=<JWT_SECRET> -e PORTAL_SECRET=<PORTAL_SECRET> -e MONGO=<MONGO> -e BASE_URL=<BASE_URL> -- npx -y @formio/uag`.
What does formio-uag do?
It gives an AI agent a Form.io form schema in readable Markdown. The agent can map conversational input to structured fields and work with submissions in agentic workflows.
Which workflows does formio-uag support?
It supports conversational input that becomes a schema-based submission, plus workflows where an agent analyses existing submission data using supplied context.
