Skip to content
Back to search

mcp-sequentialthinking-tools

🧠 An adaptation of the MCP Sequential Thinking Server to guide tool usage. This server provides recommendations for which MCP tools would be most effective at each stage.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/spences10/mcp-sequentialthinking-tools with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/spences10/mcp-sequentialthinking-tools'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/spences10--mcp-sequentialthinking-tools/badge.svg)](/listings/spences10--mcp-sequentialthinking-tools)

HTML

<a href="/listings/spences10--mcp-sequentialthinking-tools"><img src="/api/listings/spences10--mcp-sequentialthinking-tools/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md Β· README Β· 174 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # mcp-sequentialthinking-tools
  2. [![CI](https://github.com/spences10/mcp-sequentialthinking-tools/actions/workflows/ci.yaml/badge.svg)](https://github.com/spences10/mcp-sequentialthinking-tools/actions/workflows/ci.yaml)
  3. [![built with vite+](https://img.shields.io/badge/built%20with-Vite+-646CFF?logo=vite&logoColor=white)](https://viteplus.dev)
  4. [![tested with vitest](https://img.shields.io/badge/tested%20with-Vitest-6E9F18?logo=vitest&logoColor=white)](https://vitest.dev)
  5. A lightweight MCP server for recording sequential reasoning steps. It
  6. is a scratchpad with history, branching, revision metadata, and
  7. optional validation for model-authored tool plans.
  8. It does **not** discover your other MCP tools and it does **not**
  9. choose tools for the model. If you pass `available_tools` and
  10. `recommended_tools`, the server validates that the recommended names
  11. exist and stores the step.
  12. <a href="https://glama.ai/mcp/servers/zl990kfusy">
  13. <img width="380" height="200" src="https://glama.ai/mcp/servers/zl990kfusy/badge" />
  14. </a>
  15. ## Why use it?
  16. Use this when a task benefits from explicit, inspectable reasoning:
  17. - breaking a messy problem into steps;
  18. - revising or branching a plan;
  19. - keeping a small reasoning history by session;
  20. - validating tool-plan names against a supplied tool list;
  21. - clearing or inspecting reasoning history during a long agent run.
  22. Do not use it for trivial requests. It adds overhead.
  23. ## Tools
  24. ### `sequentialthinking_tools`
  25. Records one thought.
  26. Required parameters:
  27. - `thought` β€” current reasoning step
  28. - `thought_number` β€” current step number
  29. - `total_thoughts` β€” current estimate; automatically raised if lower
  30. than `thought_number`
  31. - `next_thought_needed` β€” whether another thought is needed
  32. Optional parameters:
  33. - `session_id` β€” history bucket; defaults to `default`
  34. - `is_revision`, `revises_thought`
  35. - `branch_from_thought`, `branch_id`
  36. - `needs_more_thoughts`
  37. - `available_tools` β€” array of tool names or `{ name, description }`
  38. objects
  39. - `recommended_tools` β€” model-authored recommendations to
  40. validate/store
  41. - `remaining_steps` β€” short list of upcoming steps
  42. Example:
  43. ```json
  44. {
  45. "session_id": "svelte-debug",
  46. "thought": "First inspect the route files, then run the failing check.",
  47. "thought_number": 1,
  48. "total_thoughts": 3,
  49. "next_thought_needed": true,
  50. "available_tools": ["read", "bash"],
  51. "recommended_tools": [
  52. {
  53. "tool_name": "read",
  54. "confidence": 0.9,
  55. "rationale": "Need to inspect the relevant files before editing.",
  56. "priority": 1
  57. }
  58. ]
  59. }
  60. ```
  61. If `recommended_tools` contains a name not present in
  62. `available_tools`, the call returns `isError: true` and does not store
  63. the thought.
  64. ## Security posture
  65. The server treats thought text, tool descriptions, rationales, and
  66. remaining-step text as untrusted input. Prompt-injection-like text is
  67. scanned and redacted before it is stored or returned in history. Calls
  68. with redactions include `security_warnings` showing which fields
  69. matched.
  70. This is defensive filtering, not a guarantee that arbitrary
  71. adversarial text is safe. Do not put secrets in thoughts or tool
  72. descriptions.
  73. ### `get_thinking_history`
  74. Returns stored thoughts for a session.
  75. Parameters:
  76. - `session_id` β€” defaults to `default`
  77. - `branch_id` β€” optional branch filter
  78. - `limit` β€” max records to return; default `50`, max `500`
  79. ### `clear_thinking_history`
  80. Clears one session or every session.
  81. Parameters:
  82. - `session_id` β€” defaults to `default`
  83. - `all_sessions` β€” clear all history buckets
  84. ## Prompt
  85. ### `sequential-thinking-guidance`
  86. A short prompt that tells the model how to use this server honestly:
  87. as a scratchpad and validator, not as an external reasoning engine.
  88. ## Configuration
  89. ### Claude Desktop / compatible MCP clients
  90. ```json
  91. {
  92. "mcpServers": {
  93. "mcp-sequentialthinking-tools": {
  94. "command": "npx",
  95. "args": ["-y", "mcp-sequentialthinking-tools"],
  96. "env": {
  97. "MAX_HISTORY_SIZE": "1000"
  98. }
  99. }
  100. }
  101. }
  102. ```
  103. `MAX_HISTORY_SIZE` is per session and defaults to `1000`.
  104. The server uses `tmcp` and includes a small stdio transport that
  105. accepts both standard `Content-Length` framed MCP messages and
  106. newline-delimited JSON used by older `tmcp` tooling.
  107. ## Development
  108. ```bash
  109. pnpm install
  110. pnpm test
  111. pnpm build
  112. pnpm check
  113. ```
  114. This project uses `vite-plus` for build, test, format, and lint
  115. orchestration.
  116. ## Publishing
  117. ```bash
  118. pnpm changeset
  119. pnpm changeset version
  120. pnpm release
  121. ```
  122. ## License
  123. MIT License β€” see [LICENSE](LICENSE).
  124. ## Acknowledgments
  125. - Built on the
  126. [Model Context Protocol](https://github.com/modelcontextprotocol)
  127. - Adapted from the
  128. [MCP Sequential Thinking Server](https://github.com/modelcontextprotocol/servers/blob/main/src/sequentialthinking/index.ts)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.