Models and memory · MCP Steering Group

Set up Sequential Thinking

Gives the model a scratchpad for breaking a problem into numbered thoughts it can revise or branch.

Publisher: MCP Steering Group Official
Run locally: npx -y @modelcontextprotocol/server-sequential-thinking
Transports: stdio
Authentication: None.
Environment: DISABLE_THOUGHT_LOGGING
What it can touch: No side effects outside the session.
Licence: MIT (existing) / Apache-2.0 (new contributions)
Source: github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking
Status: Active

Install

This server runs on your machine as a child process of the client. You need the runtime its command uses, such as Node.js for npx, uv for uvx or Docker, plus any variables listed under Environment.

Claude Code

shell

claude mcp add --scope project sequential-thinking -- npx -y @modelcontextprotocol/server-sequential-thinking

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "sequential-thinking": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sequential-thinking"
      ]
    }
  }
}

Codex

~/.codex/config.toml

[mcp_servers.sequential-thinking]
command = "npx"
args = ["-y", "@modelcontextprotocol/server-sequential-thinking"]

Config for other clients →

Other servers in models and memory

MCP Steering Group

Memory

Keeps a knowledge graph of entities and relations in a local file so the agent can recall facts between sessions.

Pinecone

Search Pinecone docs, and create, upsert into and query Pinecone vector indexes from a local server.

Hugging Face

Search Hub models, datasets, papers and docs, and call Gradio Spaces as tools, hosted by Hugging Face.

Sources

github.com/modelcontextprotocol/servers