The Periodix MCP server lets AI agents search, enrich, and extract data from LinkedIn and Sales Navigator using plain text. Search people, companies, posts, and jobs. Enrich any LinkedIn profile or company with full structured data. Extract post content, reactions, and comments with full text. No workflow builder, no HTTP configuration, no orchestrator. Describe what you need in natural language and the agent returns structured results.

This works with any MCP-compatible client: Claude Code, Claude Desktop, Cursor, Windsurf, or your own implementation.

- "Find DevOps engineers in Germany who work at fintech companies with 50 to 100 employees."
- "Search Sales Navigator for marketing directors at fintech companies in London with 200+ employees."
- "Find open product manager jobs at startups in New York posted in the last 7 days."
- "Find AI startups in Berlin with 50-200 employees. Get full company profiles with industry, headcount, and specializations."

- "Find LinkedIn posts about AI in outbound sales from this week. Get everyone who liked or commented on them, with their names, titles, companies, and what they commented."
- "Find John Smith, VP Sales at Stripe on LinkedIn. Pull his recent posts so I can personalize my first message."

- "List my LinkedIn connections and show me who works as VP or Director at SaaS companies."

The MCP server accepts the same types as the HTTP API: People, Companies, Posts, and Jobs search on LinkedIn Classic, plus People and Companies on Sales Navigator. On top of search, it provides full profile and company enrichment, post engagement extraction (reactions and comments including full comment text and author details), user post history with full post content, and access to your LinkedIn connection graph. You describe what you need in words, and the agent constructs the query for you.

Periodix ships aModel Context Protocol (MCP) serverso any MCP-aware agent — Claude Code, Claude Desktop, Cursor, Windsurf, and others — can call your actions as tools, with the same API key you already use for HTTP.

POST https://actions.periodix.net/mcp Authorization: Bearer <your-api-key>

The key is shown once. If you lose it, revoke it from the dashboard and create a new one.

claude mcp add periodix-actions \ --transport http \ https://actions.periodix.net/mcp \ --header "Authorization: Bearer <your-api-key>"

You should seeperiodix-actionslisted with its tools.

Add the server toclaude_desktop_config.json(location depends on your OS — see Anthropic's docs):

{ "mcpServers": { "periodix-actions": { "transport": { "type": "http", "url": "https://actions.periodix.net/mcp", "headers": { "Authorization": "Bearer <your-api-key>" } } } } }

Restart Claude Desktop. The tools appear under the 🔌 menu.

Connect from Cursor / Windsurf / custom MCP clients

Any client that supports theStreamable HTTPMCP transport works. Point it at the same URL and pass theAuthorizationheader. Refer to the client's documentation for the exact config shape.

After connecting, try this prompt in your AI agent:

"Search LinkedIn for 3 marketing managers at SaaS companies in Berlin."

If you see structured results with names, titles, companies, and profile URLs, your setup is working. Now try enrichment:

- Authentication errors— the API key is missing, mistyped, expired, or has been revoked. Create a new one in the dashboard and update your client config.
- No tools shown— make sure theAuthorizationheader is set and points athttps://actions.periodix.net/mcp.
- No LinkedIn profile connected— go to actions.periodix.net, navigate to Profiles page, and connect at least one LinkedIn account. A connected profile is required for any search or enrichment action.
- Empty search results— try broader keywords or check that the connected profile has access to the target network. Sales Navigator searches require an active Sales Navigator subscription on the connected LinkedIn profile.
- Rate limit errors— Periodix includes automatic rate limiting, but high-volume requests may be temporarily throttered. Wait a few minutes and retry.

Search global news using natural language. Webz.io News Search API returns the most relevant articles and content, with filters for source, country, language, date, sentiment, and category.

Fetch, convert, and search AWS documentation pages, with recommendations for related content.

Provides AI assistants with intelligent access to ML textbook content for creating accurate, source-grounded documentation.

Caesar is a free, keyless web search API for AI agents. Its remote MCP server exposes web_search (ranked results with citable provenance) and web_fetch (full pages as clean markdown), and works anonymously with no API key.

CatchAll is a web search API built for comprehensive event retrieval — not ranked results, but all matching records.

Provides real-time access to documentation, library popularity data, and career insights using the Serper API.

GitHits MCP gives AI agents access to open-source code search, package docs, real-world examples, dependency metadata, changelogs, and vulnerability context for better software development decisions.

Search and read PortOne documentation, including API schemas and product guides.