LinkedIn Content Planner — Claude Code MCP server: setup guide
An MCP server that turns LinkedIn drafting into an agent pipeline, reserving final review and post approval for human eyes inside a dedicated web interface.
claude mcp add --transport http linkedin-content-planner https://linkedin-mcp.theona.app/mcpWhat LinkedIn Content Planner does
This server exposes an editorial pipeline directly to AI assistants. Rather than operating social schedulers by hand, your agent calls tools such as create_post, update_post_content, and render_preview. The process mirrors code review. The assistant handles copy revisions and opens a review cycle, while team members approve or request changes before publication.
Content operations support version rollbacks and threaded comments. Inlining media as base64 strings quickly exhausts context budgets. Because of this, the server issues temporary HMAC-signed upload tickets via prepare_attachment_upload for large files. These tickets expire in 15 minutes and honor workspace storage ceilings of 250 MB.
Running the server locally demands Node 20+ and Postgres 16. Without setting ATTACHMENT_UPLOAD_SECRET across multi-instance clusters, upload tokens minted on one process will be rejected by another. Authentication handles OAuth 2.1 alongside static tokens prefixed with planner_mcp_ for non-interactive runners.
What it helps with
- Draft and iterate on LinkedIn updates straight from an agent workflow
- Review version differences and revert unwanted edits
- Upload carousel PDFs without pasting heavy base64 strings into chat
- Block automated posting until a human approves the final copy
Example prompts for the agent
- Draft a LinkedIn update explaining our migration and render a preview
- Generate an upload ticket for my slide deck and attach it to the draft
- Send this revision to human review through the planner
How to install LinkedIn Content Planner
Platform CLI: remote server
claude mcp add --transport http linkedin-content-planner https://linkedin-mcp.theona.app/mcpPlatform CLI: remote server
codex mcp add linkedin-content-planner --url https://linkedin-mcp.theona.app/mcpConfig file
{
"mcpServers": {
"linkedin-content-planner": {
"url": "https://linkedin-mcp.theona.app/mcp"
}
}
}Config file
{
"servers": {
"linkedin-content-planner": {
"type": "http",
"url": "https://linkedin-mcp.theona.app/mcp"
}
}
}Platform CLI: remote server
gemini mcp add -t http linkedin-content-planner https://linkedin-mcp.theona.app/mcpStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about LinkedIn Content Planner
How do I install LinkedIn Content Planner in Claude Code?
Run claude mcp add --transport http linkedin-content-planner https://linkedin-mcp.theona.app/mcp in your terminal. This registers the HTTP endpoint with your client.
What is the purpose of LinkedIn Content Planner?
It lets autonomous agents draft and edit LinkedIn posts while keeping humans in the loop. A person must approve the staged post in a web UI before anything goes live.
How are media attachments handled without bloating context?
Calling prepare_attachment_upload returns an HMAC-secured URL valid for 15 minutes. The agent or a script uploads the file using HTTP PUT, bypassing conversation tokens entirely.
