Generate, patch, read, and compress HTML pages for reports, dashboards, and docs with fast AI-agent speed and precise CSS selector updates.

Install

openclaw skills install fast-html-mcp

Fast HTML MCP Server

Five-tier MCP server for lightning-fast HTML generation from AI agents. 15 tools, 22 components, 25 templates — purpose-built for AI-driven page creation with sub-second patch times and AI-grade token compression.

When to Use

  • Generating complex HTML pages (reports, dashboards, landing pages, docs) from structured component specs
  • Patching existing HTML by element id or CSS selector without re-rendering the entire page
  • Compressing HTML to fit within token limits while preserving semantic content
  • Maintaining cross-section consistency across large multi-section documents
  • Building equity research, financial summaries, or data-heavy pages with charts and tables
  • Creating email newsletters, pitch decks, changelogs, or API documentation

Quick Start

npx -y @aimino.xdn/fast-html-mcp-server

Add to MCP client config:

Claude Desktop / Cursor / VS Code Copilot

{
  "mcpServers": {
    "fast-html-mcp": {
      "command": "npx",
      "args": ["-y", "@aimino.xdn/fast-html-mcp-server"]
    }
  }
}

Claude Code

claude mcp add fast-html-mcp -e npx -a "-y" -a "@aimino.xdn/fast-html-mcp-server"

Workflow

1. Discover what's available

list_templates
list_components
get_template_schema(template: "report")
get_component_schema(component: "hero")

2. Build a page

render_page(
  template: "report",
  sections: [
    {component: "hero", props: {title: "Q3 Report", badge: "Draft"}},
    {component: "data-table", props: {headers: ["Metric","Value"], rows: [["Revenue","$1.2M"]]}}
  ],
  output_path: "/tmp/report.html",
  options: {title: "Q3 Report"}
)

3. Inspect the output

read_html(path: "/tmp/report.html", mode: "compressed")

Read modes (by token efficiency):

  • structure — tree overview (70% savings)
  • content — typed blocks (74% savings)
  • compressed — summary + stats (87% savings)
  • text — plain text only (97% savings, best for token budgets)

4. Refine with surgical patches

patch_html(file_path: "/tmp/report.html", selector: "#content", html: "<p>Updated content</p>")
set_attribute(file_path: "/tmp/report.html", selector: "#main-title", attr: "class", value: "highlight")

5. Compress for token efficiency

read_html(path: "/tmp/report.html", mode: "compressed", offset: 0, limit: 1000)

6. Maintain consistency (multi-section documents)

check_consistency(path: "/tmp/report.html")
propagate_edit(path: "/tmp/report.html", entity: "Revenue", value: "$1.5M", sections: ["summary", "details"])

Resources

  • references/ — detailed examples and advanced use cases
  • scripts/ — automation scripts
  • assets/ — templates and resources