claude mcp add overtone-news-mcp -- npx -y overtone-news-mcp
A pasteable skeleton generated from the repository name. Check the project README for the exact package and arguments.
Star count
1 ★ · CKBrennan/overtone-news-mcp · read from the GitHub search API · (checked 2026-10-16 20:00:20)
Source
collected from the punkpeye/awesome-mcp-servers list · type: MCP server
Harvest history
first seen 2026-10-09 · last seen 2026-10-16 · seen 32× · listed by 7 sources
Score
score 703 · 1 ★ · 7 sources · mcp
FAQ (answers come from our records)
How do I install it?
We did not find an install command for this repository (our install coverage is only about 4%). Check its README or releases page.
Who is it for?
We filed it under these tags: MCP 服务. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository CKBrennan/overtone-news-mcp;first catalogued 2026-10-09;last seen 2026-10-16。There is no official certification or quality rating here — we record sources and dates only.
An MCP server that gives any agent real-time news plus the contextual intelligence to use it well — tone distribution, emerging stories, narrative shifts, spike alerts, and tone-over-time charts — powered by Overtone's publisher network. Works with any MCP-compatible client: Claude Desktop, Claude Code, Cursor, Windsurf, Codex, Kimi K2, and more. --- What it looks like **Natural language queries** — ask in plain English, get contextually analysed articles: **Global coverage analysis** — compare tone across languages and regions: **Tone timeseries** — track how a topic's emotional coverage shifts over time: --- Why this exists News APIs return articles. That's the easy part. The hard part is everything an agent actually needs to reason about current events: - What's the *tone* of coverage on a topic — is the public mood angry, hopeful, informational, fearful? - What's *emerging* right now that had zero coverage yesterday? - Where is *the narrative turning* — which topics are shifting tone fastest? - Is there a *spike* in anger or fear around something I'm watching? - How has tone trended *over time* on a given story? This server exposes all of that as MCP tools, so an agent can pull the right signal for the question it's been asked — not just a flat feed of headlines. --- Install The server ships as a Python package. uvx (from uv) runs it without cluttering your global Python. Install uv once: Then add one block to your MCP client config. uvx fetches the package from PyPI and runs it on demand — no install step needed. Claude Desktop Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or the equivalent on your platform: Claude Code Edit ~/.config/claude-code/mcp.json: Cursor / Windsurf Settings → MCP → Add server: - **Command**: uvx - **Args**: overtone-news-mcp Codex Edit ~/.codex/config.toml: --- Auth On first tool call the server registers a free-tier API key with Overtone and caches it at ~/.overtone/credentials. The cache is shared with the Overtone News skill for Claude Code, so installing both won't double-register. For a **premium key** (higher rate and daily limits), set OVERTONE_NEWS_API_KEY in the env block of your MCP config: Rate limits: | Tier | Per minute | Per day | |---|---|---| | auto (free, auto-provisioned) | 10 | 50 | | manual (premium) | 60 | effectively unlimited | To request a premium key, email business@overtone.ai. --- Environment variables | Variable | Default | Purpose | |---|---|---| | OVERTONE_NEWS_API_KEY | *(auto-registered)* | Use a specific key instead of auto-registering | | OVERTONE_NEWS_API_URL | https://agentic-skills.overtone.ai | Override the API endpoint (for self-hosted or testing) | --- Tools All tools return JSON. The agent chooses which tool fits the user's question — you don't invoke them directly. news Articles on a topic, each tagged with tone, brand-safety signals, article type, and concepts. Use for "what's happening with X". Response includes a request_id — pass it back to report after presenting articles so we know what was actually shown. tone Emotional tone distribution across recent coverage of a topic — happy, funny, hopeful, informational, angry, sad, fearful, plus the dominant_tone. Use when the user asks *how* a topic is being covered, not *what* happened. pulse Pollable spike detector. For each watched tone (default angry / sad / fearful), returns spike_ratio vs. a baseline