M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
A

Atom of Thoughts MCP Server

Community
Structured reasoning for LLMs: decompose, track confidence, visualize, approve.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #199 of 438
Stars
★ 92 Popular
Transport
stdio (local process)
Runtime
Node.js · Docker
Credentials
No credential needed
License
MIT
Last commit
Tools
3
54FMRS · D

This server enhances AI decision-making with structured reasoning, improving explainability. Installation is easy, supports multiple clients. Suitable for developers needing transparent reasoning, but be aware of external dependencies and trust implications.

Strongest · Documentation 13/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
9/20
Maintenance
13/20
Documentation
13/20
Setup experience
11/20
Why each score
Reliability 8/20
Evidence: README shows command usage and config flags, but no CI config or test source files (only claims 183 tests passed). Architecture seems sound, but lack of verifiable execution evidence limits score. No contradictions found, but error handling and edge cases remain unproven. Deductions: no CI configuration, no test source, no runtime verification.
Security and permissions 9/20
Evidence: Source code not visible in this static review, so credential handling and network behavior cannot be verified. README mentions a local HTTP listener (ephemeral port) for approval feedback, but security details are thin. No obvious red-line risks (e.g., credential exfiltration) are apparent, but least-privilege and confirmation mechanisms are not documented. Deductions: lack of detailed analysis of data flow, external network interactions, and file writes (e.g., approval JSON).
Maintenance 13/20
Evidence: Repository is not archived, has 73 stars and 6 open issues, but no commit history, release history, or dependency update evidence is provided. MIT license is clear. Maintenance activity is limited because recent commits or issue responses cannot be confirmed. Deductions: no commit dates, no release history, no security response channel evidence.
Documentation 13/20
Evidence: README provides installation, configuration, tool parameters, migration guide, and usage examples. However, limitations, costs, and troubleshooting are not covered. Documentation is well-structured but lacks deeper technical details and troubleshooting guidance. Deductions: no cost/limitations, no troubleshooting guide, no API reference.
Setup experience 11/20
Evidence: README provides multiple installation methods (npx, npm, Docker, Smithery) and client config examples, with clear steps. However, automated configuration verification (e.g., CI) is missing. Score is limited by static review; no real installation test was executed. Deductions: no CI evidence, no platform compatibility verification, no step-by-step verification process.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessWrites / deletes local filesUses the network

Best for

  • Developers needing transparent, auditable AI reasoning
  • Researchers decomposing complex problems into manageable subtasks
  • Users of MCP-aware clients like Claude Code, Cursor, Windsurf

Not for

  • Simple Q&A without need for structured reasoning
  • Scenarios where extra visualization or approval steps are unwelcome
  • Users who don't need depth control or atom management

Required permissions

  • Executes npx to fetch package, requires network
  • Writes temporary files for visualization HTML
  • May write to ~/Downloads for approval JSON
  • Can start a local HTTP server for browser interaction

Risks and side effects

  • Dependency on external npm packages, supply chain risk
  • Visualization may open browser, privacy consideration
  • Untrusted tool calls could produce unreliable reasoning

Setup

Before you start

Runtime：Node.js · Docker

Recommended npx: add {"command":"npx","args":["-y","@dioptx/mcp-atom-of-thoughts"]} to MCP config and restart your client. Also supports npm global, Smithery, and Docker.

.mcp.json 
{
  "mcpServers": {
    "atom-of-thoughts": {
      "command": "npx",
      "args": [
        "-y",
        "@dioptx/mcp-atom-of-thoughts"
      ]
    }
  }
}

Shown for Claude Code. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "atom-of-thoughts": {
      "command": "npx",
      "args": [
        "-y",
        "@dioptx/mcp-atom-of-thoughts"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add atom-of-thoughts -- npx -y @dioptx/mcp-atom-of-thoughts

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Restart the client and check that AoT-fast, AoT-full, and atomcommands appear in its tool list; or ask the model to call AoT-fast and confirm it returns a structured atom chain with premise, reasoning, and conclusion atoms.

Troubleshooting

  1. Ensure Node.js version meets requirements (see package.json)
  2. Check MCP config for correct command and args
  3. If visualization doesn't open, check --viz setting and browser permissions

Things to try

Once connected, you can ask your AI assistant things like:

  • Use AoT-fast to think through whether we should use JWT or session-based auth for the API.
  • Use AoT-full to decompose this architecture plan into sub-problems.
  • Use atomcommands to create a new session and then list all sessions.
  • Run AoT-fast on this problem with viz: true so I can see the dependency graph.

Tools 3

AoT-fast writes
Quick reasoning for tradeoffs, debugging, decisions, option evaluation (depth 3).
AoT-full writes
In-depth reasoning for plans, architecture, decomposition (depth 5).
atomcommands writes
Manage sessions, export graphs, poll approvals, and lifecycle.

Use cases

Use AoT-fast for quick tech decisions (e.g., JWT vs session auth).
Use AoT-full for complex planning or architecture decomposition.
Visualize reasoning via TUI or browser and approve/reject atoms.

Supported clients

Claude Code
Cursor
Windsurf

Listed from the project's documentation, not tested by this site.

Overview

Atom of Thoughts is an MCP server for structured AI reasoning. It decomposes problems into atomic thoughts (premise, reasoning, hypothesis, verification, conclusion), each tagged with a confidence score, forming a dependency chain. The server offers multiple tools (AoT-fast, AoT-full, atomcommands) to build and manage thought chains, and supports interactive visualization via a terminal TUI or browser (D3.js), allowing users to approve or reject atoms for human oversight. It is based on the paper 'Atom of Thoughts' (arXiv:2502.12018), written in TypeScript, and can be installed via npx, npm, Smithery, or Docker.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 9df1eca138fa Data synced 2026-10-11 Read the FMRS scoring method