Perigon MCP Server Overview

Perigon’s MCP server bridges the gap between AI assistants and real‑time news data. By exposing the Perigon News API through the Model Context Protocol, developers can give Claude and other agents instant access to up‑to‑date headlines, journalist profiles, and company coverage without building custom connectors. This solves a common pain point for data‑centric workflows: the need to fetch, parse, and format news feeds each time an assistant needs fresh information.

The server accepts authenticated requests over both HTTP (streamable) and Server‑Sent Events (SSE). Clients register a single endpoint () and supply a Perigon API key in the header. Once connected, agents can issue simple JSON queries that specify topics, geographic focus, time ranges, or source types. The MCP server translates these into Perigon API calls and streams back structured results, preserving low latency while keeping the client code minimal.

Key capabilities include:

• Dynamic news retrieval: request top headlines, trending stories, or niche coverage (e.g., climate policy journalists) for any region and time window.
• Journalist & source discovery: find reporters by beat, publication, or location and pull their recent articles.
• Entity tracking: search for companies or people (e.g., Tesla CEOs) and receive the latest media mentions.
• Rate‑limit awareness: built‑in handling of Perigon’s limits ensures smooth operation even under heavy query loads.

Typical use cases span journalism automation, market research bots, and AI‑powered briefing tools. For example, a newsroom assistant can pull the latest political headlines for a live broadcast, while a financial analyst agent can gather recent coverage of specific companies before generating a risk report. The server’s simple, standard‑compliant interface means it plugs into any MCP‑compatible client—whether a local tool, cloud service, or the Smithery platform.

Because it exposes the full breadth of Perigon’s data through a single, well‑documented endpoint, developers gain rapid, reliable access to high‑quality news without managing API keys or parsing logic. The result is a streamlined workflow where AI assistants can focus on analysis and synthesis, while the MCP server handles data acquisition and formatting behind the scenes.