Docs/For AI Assistants/MCP Server

    MCP Server

    Let your own AI assistant work your Planino pipeline.

    What It Is

    Planino runs an MCP server, the standard way AI assistants such as Claude Code or Cursor connect to other tools. Once connected, your assistant can read your pipeline, file ideas, prepare posts and attach media for you.

    Creating a Token

    Open Settings > MCP Server, give the token a name and create it. Copy the token right away: Planino shows it only once. You can revoke a token on the same page at any time.

    A token acts as you. Keep it out of public places such as shared documents or code repositories.

    Connecting Your Assistant

    Add the server to your assistant's MCP configuration, with your token in the Authorization header:

    json
    {
      "mcpServers": {
        "planino": {
          "type": "http",
          "url": "https://planino.studio/api/public/mcp",
          "headers": { "Authorization": "Bearer pln_mcp_..." }
        }
      }
    }

    The MCP Server page in Settings shows the same snippet with a copy button.

    What Your Assistant Can Do

    • Publishing profiles: list_publishing_profiles, create_publishing_profile, update_publishing_profile, create_language_edition
    • Ideas and projects: create_idea, list_ideas, get_idea, evolve_idea, update_project, list_content, get_content, get_pipeline_overview
    • Posts: check_post_media, create_posts, update_post, list_posts, mark_posted
    • Media: begin_media_upload, finish_media_upload, abort_media_upload, attach_media, list_media, media_usage, delete_media
    • Transcripts and subtitles: transcribe_media, get_transcript, set_transcript, search_transcripts, get_subtitles, set_subtitles, link_subtitles
    • Pictures and thumbnails: render_visual, refine_visual, list_visuals, set_thumbnail_brief, render_thumbnail, refine_thumbnail, derive_thumbnail_vertical, list_thumbnails
    • AI poster jobs: list_browser_jobs, claim_browser_job, report_browser_job
    • YouTube channel: get_channel_stats, set_channel_keywords, set_channel_description, set_video_description

    Each tool describes its own inputs to the assistant, so you can ask in plain words, for example "file this as an idea" or "make a LinkedIn post from project 25".