← all integrations

integrations/sequential-thinking-mcp

Sequential Thinking MCP mcp

A scratchpad tool the model calls to write its reasoning as numbered, revisable steps; it reaches nothing outside itself.

Third-party software. Forgeprint hosts none of this code: it is an installation recipe for a project somebody else publishes and maintains. Review the permissions below before installing.

Upstream

What it can reach

Keeps the thoughts the model sends it in memory for the session and returns a count; it reads and writes no file, network or account.

It needs no secret.

Install

Claude Code

claude mcp add-json sequential-thinking '{"command":"npx","args":["-y","@modelcontextprotocol/server-sequential-thinking@2026.8.31"],"env":{"DISABLE_THOUGHT_LOGGING":"true"}}'

Codex CLI

codex mcp add sequential-thinking --env DISABLE_THOUGHT_LOGGING=true -- npx -y @modelcontextprotocol/server-sequential-thinking@2026.8.31

Gemini CLI

gemini mcp add sequential-thinking -e DISABLE_THOUGHT_LOGGING=true npx -y @modelcontextprotocol/server-sequential-thinking@2026.8.31

Only agents whose command syntax has been verified are listed. A missing agent is one nobody has confirmed, not one that does not work.

Read it

README.md