Sequential Thinking MCP Server: Setup for Claude, Cursor and Other Apps
@modelcontextprotocol/server-sequential-thinkingmodelcontextprotocol/servers/tree/HEAD/src/sequentialthinkingGives Claude, Cursor and other assistants a scratchpad for step-by-step reasoning that can revise and branch.
Sequential Thinking is a small server from the Model Context Protocol project. It gives your assistant one tool for working through a problem in numbered thoughts, which it can revise or branch as it goes. It runs on your computer and needs no account or API key. The tool records thoughts and returns counts, so the quality of the reasoning still depends on your assistant.
Good for
- People who want an assistant to plan or debug in explicit steps
- Developers working through designs where the scope is unclear at first
- Anyone who wants to see the reasoning, not only the answer
Works in
12 should work
Tap an app to see how to set it up and any known problems.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Supports local MCP servers.
Client does not support this transport.
Client does not support this transport.
Install sequentialthinking MCP
Claude Code
- 1
Copy this prompt.
Add the sequentialthinking MCP server to Claude Code. Run this command: claude mcp add sequentialthinking -- npx -y @modelcontextprotocol/server-sequential-thinking Then check that it loaded and tell me if I need to sign in or restart.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Requires Node.js.
Built from the package data we hold. Not yet tested by us.
Tools
Lets the assistant reason in numbered steps and revise or branch them
The assistant records each thought with a number and says whether it needs more. It can mark a thought as a revision of an earlier one, or branch to try another path. The server keeps the history while the connection lasts and returns the counts.
Try: “Plan a database migration from PostgreSQL 14 to 16, list the risks, and revise the plan if downtime would exceed 5 minutes.”
Write ·
sequentialthinking
Read from the source code
Permissions
low risk · 1 thing to watch
Low risk
It only keeps the assistant's thoughts in memory for the session. No special precautions are needed.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
Thoughts are printed to the server log
Each thought is written to the server's error output unless you set DISABLE_THOUGHT_LOGGING to true. Set it if your thoughts may contain private details. github.com
Security advisories
No published security advisories as of Nov 1, 2026.
Technical details
| Turning on read-only | No read-only switch is documented. The tool only adds to an in-memory thought history. |
|---|---|
| Sign-in and keys | No credentials needed. |
Questions
Does Sequential Thinking make the AI smarter?
It gives the assistant a place to lay out and revise its reasoning in steps. The tool records thoughts and returns counts, so it helps structure the work but adds no new knowledge.
Does it need an API key?
No. It runs on your computer with no account, sign-in or API key.
How do I know it is working?
Ask a multi-step question and watch for repeated calls to the tool in your app's tool activity. The README suggests confirming the tool appears in the host's tool list first.
Can I turn off logging of my thoughts?
Yes. Set the DISABLE_THOUGHT_LOGGING variable to true. Thoughts are otherwise printed to the server's error output.
What do I need to run it?
Node.js 24 or later when it runs through npx, or Docker if you use the image the README documents.
Does it remember thoughts between chats?
No. The README says the thought history lasts as long as the client's connection.
Information
updated Nov 2, 2026
- Publisher
- Community
- Package
@modelcontextprotocol/server-sequential-thinking- Registry
- npm
- Transport
- stdio
- Auth
- None
- License
- MIT
- Last commit
- Oct 29, 2026
- Open issues
- 26
- Listed in
- Docker MCP Catalog
Sources
Docs, issue trackers and discussions this page is built from. Issues, quotes and findings above link to their own source.
Show all 2Hide
- Sequential Thinking MCP Server README · github.com
- Sequential Thinking server source · github.com