For AI clients

MCP server

An assistant can ask this site directly which PDF tool fits a job and get back the link that opens it, instead of guessing from a search result.

The endpoint

https://pdf.novusstreamsolutions.com/mcp

One stateless POST, JSON-RPC over HTTP, protocol revision 2026-07-28. No account, no key, no sign-up: every tool reads content already public on this site, so an authentication step would be theatre. A GET answers 405 on purpose, because that revision removed the server-sent stream and a client opening one needs a definitive answer rather than a request that hangs.

Adding it to a client

Most clients take a block like this. The server name is yours to choose; the URL is the part that matters.

{
  "mcpServers": {
    "novus-pdf": {
      "type": "http",
      "url": "https://pdf.novusstreamsolutions.com/mcp"
    }
  }
}

What it exposes

List every tool and workspace on this sitelist_capabilities: Every single-purpose PDF tool this site offers plus the three workspaces that are not tools (the form editor, the signing page and the blank-page builder) with what each accepts, what it produces, and the URL that opens it. Use this to answer "can this site do X?" and to pick the right surface before sending someone to it. These all run in the visitor's browser, so this server can describe and locate them but cannot execute them.
Find the right tool for a described jobfind_tool: Search this site's tools and workspaces by what someone is trying to do, in their own words, "put two contracts into one file", "my scan is sideways", "add a signature to a form", "make this small enough to email". Returns the best matches with the URL that opens each one. This server cannot run the tool; the URL is the handoff.
Get the URL and steps to run one toolopen_tool: The URL that opens one tool or workspace, plus the authored steps to operate it and the limitations it states up front. This is the handoff point: this server cannot process a document (everything runs client-side and nothing is uploaded), so the caller (an AI browser, or the person) opens this URL and drives the page. Call list_capabilities or find_tool first to get an id.
Search the guides, tutorials, FAQ and articlessearch_docs: Search this site's written material: the guides, tutorials, tips, FAQ answers, how-it-works steps, blog articles, and the product and policy pages. Use it for a how-to or a policy question rather than a which-tool question, find_tool is the better door for that, and to cite a page. Deep links carry the anchor of the section they name.
List every site in the Novus networklist_novus_sites: The other Novus Stream Solutions sites and what each one is for. Use this when a request is outside what this site does. Converting a file that is not a PDF, editing an image, study material, a music video, AI usage stats, food-market data, or a game. Each site runs its own MCP server at /mcp.

What it will not do

It cannot edit a PDF, or open your file at all. Every tool here runs inside your own browser tab on a document you chose, and nothing is uploaded, so there is no server-side pipeline for a tool to call. A tool that offered to merge or sign a document would either have to ask for that upload or lie about the result.

What it returns is which tool fits, what that tool will and will not preserve, and the URL that opens it, so the browser doing the work can be an AI browser or your own. The same limits a person meets are documented on Features.