Skip to content
botshub
Visit Model Salt

MCP server that gives your model randomness from outside it, so one prompt gives different answers that still do what you asked.

About Model Salt

Model Salt is an MCP server that gives a language model randomness from outside the model, so the same prompt can produce different answers that still do what was asked. The README explains that the model picks out the main open choices in a request, such as style, layout or names, and external randomness chooses one option for each before the model completes the task.

The README describes a single model_salt tool with a variety setting from 0 to 2 and a field for keeping choices from an earlier run. The model skips it for questions with one right answer. It needs Node.js 18 or later. Claude Code or Codex, if signed in, can run the extra step on the user's own plan, otherwise the model does it itself using more tokens. Setup is a one-time npx command for clients such as Claude Code, Codex, Gemini CLI or Kimi Code,. Code is MIT licensed. It is maintained by Chris Benjaminsen. Details are from the README.

Key features

  • External randomness for varied answers
  • Variety setting from 0 to 2
  • Keep field for liked choices
  • Setup for several coding clients
  • MIT code licence

More ai & ml services MCP servers

  • Blender MCP

    ahujasid

    AI & ML services

    Community plugin to control Blender 3D with any LLM of your choice

    34K GitHub starsPythonMIT
  • AI & ML services

    The markitdown-mcp package provides a lightweight STDIO, Streamable HTTP, and SSE MCP server for calling MarkItDown.

    196K GitHub starsPythonMIT
  • Memory MCP Server

    modelcontextprotocol

    AI & ML services

    A basic implementation of persistent memory using a local knowledge graph. This lets Claude remember information about the user across chats.

    92K GitHub stars
  • Sequential Thinking MCP Server

    modelcontextprotocol

    AI & ML services

    An MCP server implementation that provides a tool for dynamic and reflective problem-solving through a structured thinking process.

    92K GitHub stars
  • AI & ML services

    It's like v0, but in your Cursor / Claude Code / Windsurf: search 10,000+ React/Tailwind components, generate new UI with AI, and publish your own — right from your editor. Magic MCP is now the 21st MCP; this package keeps old configs working. Setup: 21st.dev/mcp

    6.6K GitHub starsTypeScriptISC
  • AI & ML services

    A visualization mcp & skills contains 25+ visual charts using @antvis. Using for chart generation and data analysis.

    4.4K GitHub starsTypeScriptMIT
Report this listing
Is this your project? Claim this listing