delta-mcp-prototype
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@delta-mcp-prototype search for mentions of Tesla from last week"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
delta-mcp-prototype
An MCP (Model Context Protocol) server that wraps the UMT media monitoring API, exposing media mention search as a tool for MCP-compatible clients (e.g. Claude Desktop, Claude Code).
Requirements
Node.js 22 (see
.nvmrc)A UMT account (username/password)
Related MCP server: news-monitor-mcp
Installation
npm installUsage
Run the server with your UMT credentials passed as environment variables:
UMT_USERNAME=<username> UMT_PASSWORD=<password> node index.js [--env delta|beta]UMT_USERNAME/UMT_PASSWORD— UMT account credentials (required). Environment variables are recommended over CLI flags since flags are visible to other processes/users via the process list;--username/--passwordare still accepted as a fallback for quick local testing.--env— API environment,delta(default) orbeta
The server communicates over stdio, so it's meant to be launched by an MCP client rather than run standalone.
Connecting to an MCP client
Example configuration (e.g. Claude Desktop claude_desktop_config.json):
{
"mcpServers": {
"delta-mcp-prototype": {
"command": "npx",
"args": ["delta-mcp-prototype", "--env", "beta"],
"env": {
"UMT_USERNAME": "<username>",
"UMT_PASSWORD": "<password>"
}
}
}
}Tools
fetch_mentions— Fetch media mentions with filters for search id, phrase, date range, language, country, media segment, sentiment, and tags. Returns a paginated list of mentions (title, url, media segment, published date, sentiment, virality, link counts).current_date_and_time— Returns the current date, time, and timezone; useful for resolving relative date ranges.
Build
Bundle the server into a single file with esbuild:
npm run buildOutput is written to dist/index.js.
Project structure
index.js CLI entry point — parses args, starts the server
src/index.js MCP server setup and tool registration
src/tools.js Tool definitions (fetch_mentions, current_date_and_time)
src/umt-dao.js UMT API client (auth + mentions endpoint)
src/env.js Environment/base URL constants
src/utils.js Small helpers (object path access, TOON encoding, try/catch)Available Tools
2 toolscurrent_date_and_timecurrent_date_and_timeA
Returns the current date and time. Call this whenever you need to know today's date or the current time, for example when the user asks 'what day is it?' or when computing relative date ranges like 'last 7 days'.
Inputs: no named parameters.
fetch_mentionsfetch_mentionsA
Fetch media mentions matching the specified filters. Supports filtering by search, phrase, date range, language, country, media segment, sentiment, tags, source, read/marked/critical/mailed status, inlink count, and more. Results are paginated and sortable. Returns total count and a list of mentions with title, url, media segment, published date, sentiment, virality, and link counts.
Inputs: limit, offset, phrase, country, +7 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
The two tools serve completely different purposes: one provides current date/time, the other fetches media mentions with many filters. There is no overlap or ambiguity.
Both tool names follow a consistent verb_noun pattern using snake_case ('current_date_and_time', 'fetch_mentions'), making them predictable and easy to understand.
With only two tools, the server feels thin for a media monitoring prototype. The inclusion of a generic date/time utility alongside a specialized mentions fetcher suggests a lack of focused scope, but it is not extreme.
The fetch_mentions tool is comprehensive with many filters, but there are no tools for creating, updating, or managing mentions. The date/time tool is unrelated to the core domain, leaving significant gaps for a full media monitoring workflow.
Related MCP Connectors
OAuth-enabled MCP server that extracts structured data from public social media posts.
MCP server for Altmetric APIs - track research attention across news, policy, social media, and more
Remote MCP server that returns sentiment analysis results from SentinelScan API.
One MCP server for all web data. Scrape, search, crawl, and AI research in one install.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA MCP Server Implementation that integrates the Marketaux api providing search based on entity, countries, industries, symbols etc.5ISC
- AlicenseAqualityBmaintenanceMCP server for global news monitoring, media analysis and sentiment tracking via WorldNewsAPI — full-text search across 150+ countries, German/English sentiment analysis, top headlines, GL briefings, newspaper front pages and geo-search. API key required.1525 PyPIMIT
- FlicenseNot gradedqualityDmaintenanceMCP server providing web search, news search, and X/Twitter search capabilities via HTTP or stdio.-
- AlicenseNot gradedqualityDmaintenanceA remote, read-only MCP server for global media intelligence using GDELT DOC 2.0 API, enabling search and analysis of multilingual news coverage, tone, and attention trends.MIT