MCP Automations

wzltmp/mcp-automations

A production-grade MCP server that connects Claude to four content automation tasks: URL summarization with trafilatura, content repurposing into threads or newsletters, Tavily-powered news digests, and competitor discovery. Every tool returns typed Pydantic models with per-call token counts and dollar costs attached, routing cheap tasks to Haiku and writing-heavy ones to Sonnet. Ships with both stdio and HTTP transports, so the same codebase serves Claude Desktop locally and remote clients over HTTPS. Includes a Streamlit playground with per-session spend caps and a working Fly.io deployment. Reach for this when you want to see MCP's three primitives (tools, resources, prompts) implemented together with proper cost telemetry, not just the happy path.

tools: 4
transport: HTTP
last updated: May 28, 2026

View on GitHub

Install

From registry metadata

$ npx add-mcp https://mcp-automations.fly.dev/mcp --name mcp-automations

Run in your terminal. add-mcp asks which agents to add it to (Claude Code, Codex, Cursor and more). Add -g to install it for every project.

Review the command, arguments and environment values before installing. MCP servers run with your local permissions.

Other ways

Claude Code

$ claude mcp add --transport http mcp-automations https://mcp-automations.fly.dev/mcp

Run in your terminal. Add --scope user to make it available in every project.

Tools

Verified live · 4 tools

Verified live against the running server on Nov 6, 2026.

summarize_urlFetch a URL, extract clean article text, and return an N-bullet summary.2 params

Fetch a URL, extract clean article text, and return an N-bullet summary.

Parameters* required

n_bulletsinteger

default: 5

url*string

repurpose_contentRepurpose long-form text into ``twitter_thread``, ``linkedin_post``, or ``newsletter``.2 params

Repurpose long-form text into ``twitter_thread``, ``linkedin_post``, or ``newsletter``.

Parameters* required

formatstring

default: twitter_thread

text*string

daily_digestSearch the web for recent news on ``topic`` and return a digest with citations.2 params

Search the web for recent news on ``topic`` and return a digest with citations.

Parameters* required

n_resultsinteger

default: 5

topic*string

find_competitorsIdentify ``n`` plausible competitors for a company at ``domain`` (e.g., 'stripe.com').2 params

Identify ``n`` plausible competitors for a company at ``domain`` (e.g., 'stripe.com').

Parameters* required

domain*string

ninteger

default: 5

Readme

MCP Automations

A production-grade Model Context Protocol server in Python — four LLM-callable tools, two transports, deployed two different ways.

	URL
Source	https://github.com/wzltmp/mcp-automations
Playground (browser demo)	https://mcp-automations-5vgea2ynuyrvbzkcxm6yoh.streamlit.app/
MCP HTTP server	https://mcp-automations.fly.dev/mcp

# 30-second proof the server is up:
curl -X POST https://mcp-automations.fly.dev/mcp \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","method":"initialize","id":1,
       "params":{"protocolVersion":"2024-11-05",
                 "capabilities":{},
                 "clientInfo":{"name":"curl","version":"1"}}}'

What this is

Most "AI engineer" portfolio projects are applications (a RAG chatbot, an agent that does research). This project is the layer underneath — the typed tools an LLM can call and the transport plumbing that exposes them. MCP is the emerging standard for LLM tool use (~97M monthly SDK downloads as of early 2026); building one — not just consuming one — is the rare skill.

View the full README on GitHub

Details

Repository: github.com/wzltmp/mcp-automations
Website: github.com/wzltmp/mcp-automations
Transport: HTTP
Registry: MCP registry, active
Resources: 1
Prompts: 1
Tools verified: Nov 6, 2026