Agent Quickstart
Linkedmash gives an AI agent a tool surface to manage a user's saved LinkedIn posts and to create, schedule, publish, and analyze LinkedIn content — over MCP or REST.
https://api.linkedmash.com/v1 directly. Both paths use the same token and the same account.Connect in four steps
Get an API token
Generate a token in your Linkedmash account. The agent sends it on every request as Authorization: Bearer <token>. Treat it like a password — never commit it or share it.
Connect the agent
Pick how the agent talks to Linkedmash. MCP is recommended — your agent gets typed tools with no HTTP plumbing.
Option A — MCP (recommended)
Follow Connect Your Agent for copy-paste config for Claude Desktop, Claude Code, Cursor, VS Code, and more — the agent gets all 22 tools automatically. The hosted server is https://mcp.linkedmash.com/api/mcp with your API token in the URL.
Option B — REST
Call https://api.linkedmash.com/v1 directly with a bearer token. Use this when your agent already speaks HTTP or MCP isn't available.
Authorization: Bearer YOUR_API_TOKENKnow the capabilities
Saved posts work on any plan. Content Studio, Scheduling, and Analytics require a paid Creator or Lifetime plan.
Saved posts
Content Studio (paid)
Scheduling (paid)
Analytics (paid)
Install the skill and try it
Install the Linkedmash skill in your agent — npx skills add linkedmash/agent-skills (Claude Code, Cursor, …) — so it knows the tools and workflows, then run an example.
First request
List the most recent saved posts for the authenticated user:
curl -X GET "https://api.linkedmash.com/v1/bookmarks?limit=5" \
-H "Authorization: Bearer YOUR_API_TOKEN"Over MCP, the same call is just the list_bookmarks tool — no URL or headers to assemble:
https://mcp.linkedmash.com/api/mcp?token=YOUR_API_TOKEN