Skip to main content
Glama
maxylev

http-mcp-server

http-mcp-server

A minimal Model Context Protocol (MCP) HTTP server implementing the 2026-07-28 protocol, including the Tasks extension. Zero runtime dependencies.

Usage

Run directly with npx (requires Node.js 24+):

npx http-mcp-server

Or install and run:

npm install -g http-mcp-server
http-mcp-server

The server listens on http://127.0.0.1:8456/mcp.

Related MCP server: MCP HTTP Server Demo

Configuration

Variable

Default

Description

PORT

8456

Listening port

HOST

127.0.0.1

Listening interface

Endpoints

  • POST /mcp — MCP JSON-RPC over HTTP (2026-07-28)

  • POST /mcp-no-tools — same, but tools/list returns no tools (for testing empty-server behavior)

  • GET /mcp — SSE stream

Protocol

  • Discovery: server/discover advertises supported protocol versions and the io.modelcontextprotocol/tasks extension capability. Legacy initialize is also supported.

  • Headers: requests carrying MCP-Protocol-Version, Mcp-Method, and Mcp-Name headers are validated against the body (-32020 on mismatch, -32022 for unsupported versions). Requests without the new-style headers are accepted for backward compatibility.

  • Tasks: clients that declare io.modelcontextprotocol/tasks in their per-request _meta.io.modelcontextprotocol/clientCapabilities.extensions receive a task handle (resultType: "task") from thx_slow_echo and poll tasks/get until a terminal state, answering elicitations via tasks/update and cancelling via tasks/cancel.

Tools

Tool

Description

thx_echo

Echoes the given text back as echo:<text>

thx_add

Sums numbers as sum:<n>

thx_upper

Uppercases text as upper:<TEXT>

thx_time

Returns the current UTC time as time:<ISO 8601>

thx_slow_echo

Async task tool: echoes after delayMs; optionally pauses for an approval elicitation (requireApproval)

thx_slow_echo example

{
  "name": "thx_slow_echo",
  "arguments": { "text": "hello", "delayMs": 1000, "requireApproval": false }
}

The call returns a task handle:

{
  "result": {
    "resultType": "task",
    "taskId": "9b1deb4d-3b7d-4bad-9bdd-2b0d7b3dcb6d",
    "status": "working",
    "statusMessage": "echoing 'hello' in 1000ms",
    "createdAt": "2026-08-29T10:30:00Z",
    "lastUpdatedAt": "2026-08-29T10:30:00Z",
    "ttlMs": 3600000,
    "pollIntervalMs": 100
  }
}

Poll tasks/get until the task reaches completed, failed, or cancelled. With requireApproval: true the task enters input_required and surfaces an elicitation/create request under inputRequests; answer it with tasks/update and inputResponses.

Development

npm install
npm test        # run tests
npm run lint    # run eslint
npm run format  # format with prettier

License

MIT

A
license - permissive license
Not graded
quality - not tested
B
maintenance

Maintenance

Maintainers
Response time
0dRelease cycle
2Releases (12mo)
Commit activity

Related MCP Servers

  • F
    license
    Not graded
    quality
    D
    maintenance
    A minimal Model Context Protocol (MCP) server that uses streamable HTTP transport to provide demo tools for calculations, notes, and time. It serves as a standalone example for testing MCP connectivity and gateway registration through a standard HTTP endpoint.
  • A
    license
    Not graded
    quality
    D
    maintenance
    Exposes any stdio-based MCP server to the internet via HTTP/SSE transport, enabling remote agents to access MCP tools over a network.
    21
    MIT
  • A
    license
    Not graded
    quality
    D
    maintenance
    Enables MCP client interaction via streamable HTTP, providing example tools (echo, getPostsByUser) and resources (posts, users) with pluggable authentication providers.
    6
    MIT

View all related MCP servers

Related MCP Connectors

  • An authenticated remote MCP server for user-owned devices and one-shot capability invocation.

  • AI Reasoning Cache & Consensus Layer with 11 MCP tools via Streamable HTTP.

  • Hosted AgentLux MCP server for marketplace, identity, creator, services, and social flows.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/maxylev/http-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server