MAXential Thinking MCP server
✓ VerifiedProvides a persistent reasoning workspace for AI to build thought chains, explore alternative paths through branching, and manage complex reasoning sessions.
Also known as BAM-DevCrew/MAXential-Thinking-MCP, @bam-devcrew/maxential-thinking-mcp
- Stars
- 3
- Tools
- 20
- Downloads, 30 days
- 896
- Latest version
- 2.0.1
- Last commit
- May 25, 2026
- License
- MIT
What is MAXential Thinking MCP?
MAXential Thinking provides a persistent reasoning workspace for AI to build thought chains, explore alternative paths through branching, and manage complex reasoning sessions. It exposes 20 tools, such as think, revise, complete.
Summary generated from the project's README
About MAXential Thinking MCP server
MAXential-Thinking-MCP provides a structured workspace for AI models to manage complex reasoning tasks. Unlike standard ephemeral reasoning that disappears when context windows fill, this server uses a SQLite database to persist thought chains, branches, and tags. It is designed for developers using Claude Desktop or Claude Code who need to maintain continuity across server restarts or new conversations.
The system offers 20 specialized tools categorized into core thinking, branching, navigation, organization, and session persistence. Users can build thought chains with the think tool, explore alternative paths using branch and switch_branch, or revise previous ideas with the revise tool. The server also supports advanced organization through semantic tagging and searching, as well as visualization via Mermaid diagrams or ASCII structures. By utilizing session_save and session_load, an AI can resume deep analytical work exactly where it left off, making it ideal for long-term problem solving and architectural decision making.
What you can do with it
- 1Explore multiple technical approachesAn AI can create separate branches to evaluate different technologies like JWT or OAuth independently before merging the best insights.
- 2Maintain reasoning across sessionsYou can instruct an AI to restore a previous analytical session to continue working on a complex problem after a context reset.
- 3Organize and retrieve specific insightsA developer can ask the AI to find all previous reasoning steps that were marked with a specific semantic tag.
- 4Visualize complex thought structuresThe server can generate diagrams to show how different reasoning paths and branches connect to one another.
Written from the project's README. Every use case is checked against a quote from it.
How to install MAXential Thinking MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall MAXential Thinking MCP in Claude Code
$ claude mcp add maxential-thinking-mcp -- npx -y @bam-devcrew/maxential-thinking-mcpInstall MAXential Thinking MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"maxential-thinking-mcp": {
"command": "npx",
"args": [
"-y",
"@bam-devcrew/maxential-thinking-mcp"
]
}
}
}Install MAXential Thinking MCP in VS Code
.vscode/mcp.json
{
"servers": {
"maxential-thinking-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@bam-devcrew/maxential-thinking-mcp"
]
}
}
}Install MAXential Thinking MCP in VS Code CLI
$ code --add-mcp '{"name":"maxential-thinking-mcp","type":"stdio","command":"npx","args":["-y","@bam-devcrew/maxential-thinking-mcp"]}'Install MAXential Thinking MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"maxential-thinking-mcp": {
"command": "npx",
"args": [
"-y",
"@bam-devcrew/maxential-thinking-mcp"
]
}
}
}Install MAXential Thinking MCP in Codex
$ codex mcp add maxential-thinking-mcp -- npx -y @bam-devcrew/maxential-thinking-mcpInstall MAXential Thinking MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.maxential-thinking-mcp] command = "npx" args = ["-y", "@bam-devcrew/maxential-thinking-mcp"]
Install MAXential Thinking MCP in Gemini CLI
$ gemini mcp add -s user maxential-thinking-mcp npx -- -y @bam-devcrew/maxential-thinking-mcpInstall MAXential Thinking MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"maxential-thinking-mcp": {
"command": "npx",
"args": [
"-y",
"@bam-devcrew/maxential-thinking-mcp"
]
}
}
}MAXential Thinking MCP tools 20
Detected from the READMEthinkrevisecompleteresetbranchswitch_branchlist_branchesget_branchclose_branchmerge_branchget_thoughtget_history
Show all 20 toolsShow fewer tools
tagsearchexportvisualizesession_savesession_loadsession_listsession_summary
Configuration
MAXENTIAL_DB_PATHMAXENTIAL_LOG_FILE
Automated checks
Ran Nov 14, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachablebam-devcrew/maxential-thinking-mcp is public and reachablegithub.com/bam-devcrew/maxential-thinking-mcp ↗Pass
- Package names this repository as its source@bam-devcrew/maxential-thinking-mcp names bam-devcrew/maxential-thinking-mcp as sourcenpmjs.com/package/@bam-devcrew/maxential-thinking-mcp ↗Pass
- Build provenance (Sigstore attestation)No build attestation published for this releaseNot checked
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-05-25github.com/bam-devcrew/maxential-thinking-mcp ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from @bam-devcrew/maxential-thinking-mcpPass
- No critical advisory in the package or its dependencies135 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model20 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNo tool named like delete, drop or execPass
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubMAXential Thinking MCP
A structured, persistent reasoning workspace for AI. Provides 20 tools for building thought chains, exploring alternatives through branching, revising earlier thinking, searching reasoning history, and persisting sessions across context window resets and server restarts.
AI's built-in reasoning is ephemeral — when context windows fill, thinking gets compressed or lost. Complex problems need exploration of multiple approaches, backtracking when paths fail, and the ability to resume where you left off. MAXential externalizes reasoning into a workspace that persists, branches, and survives.
Origin
Forked from Anthropic's sequential-thinking MCP server, which provided a single tool with 9 parameters. Its schema included branching parameters (branch_from_thought, branch_id) but had no tools to create, switch, or manage branches — the parameters were effectively inert.
MAXential replaced that single tool entirely and built 20 purpose-specific tools across three releases:
| Version | What was built |
|---|---|
| v2.0 | Replaced the single tool with 11 focused tools: core thinking (think, revise, complete), full branch management (branch, switch_branch, list_branches, get_branch, close_branch, merge_branch), and navigation (get_thought, get_history) |
| v2.2 | Added 5 organization tools: tag, search, export, visualize, reset |
| v2.3 | Added 4 session persistence tools with SQLite storage: session_save, session_load, session_list, session_summary |
Tools
Core Thinking
| Tool | What it does |
|---|---|
think |
Add a thought to the reasoning chain. Thoughts are numbered and persisted automatically. |
revise |
Revise a previous thought when earlier thinking was flawed or incomplete. The original is preserved with revision history. |
complete |
Mark the thinking chain complete with a final conclusion. |
reset |
Clear the current session and start fresh. |
Branching
| Tool | What it does |
|---|---|
branch |
Create a new reasoning branch to explore an alternative path without losing the main thread. |
switch_branch |
Switch context to a different branch, or back to main. |
list_branches |
List all branches with their status and thought counts. |
get_branch |
Retrieve complete details of a specific branch. |
close_branch |
Close a branch with an optional conclusion. |
merge_branch |
Merge insights from a branch back into main. Strategies: conclusion_only, full_integration, summary. |
Navigation
| Tool | What it does |
|---|---|
get_thought |
Retrieve a specific thought by its number. |
get_history |
Get thought history, optionally filtered by branch. |
Organization
| Tool | What it does |
|---|---|
tag |
Add or remove semantic tags on a thought (e.g., hypothesis, evidence, decision, finding). |
search |
Search thoughts by content text or by tags. |
export |
Export the thinking chain as markdown or JSON. |
visualize |
Generate ASCII or Mermaid diagrams of the thought structure and branches. |
Session Persistence
| Tool | What it does |
|---|---|
session_save |
Name and describe the current session for later retrieval. |
session_load |
Restore a saved session — all thoughts, branches, and tags are loaded back into memory. |
session_list |
Browse available sessions, most recently updated first. |
session_summary |
Generate a compressed summary of a session for token-efficient context loading. |
Sessions are automatically persisted to SQLite as you work. Every think, branch, tag, and revise call writes through to disk in real time. Sessions survive server restarts, context window resets, and new conversations — pick up where you left off.
What the original provides vs. what MAXential provides
| Capability | Anthropic sequential-thinking | MAXential Thinking |
|---|---|---|
| Interface | 1 tool, 9 parameters | 20 focused tools |
| Branching | Parameters in schema, no implementation | Full lifecycle: create, switch, list, inspect, merge, close |
| Revision | Not supported | Revise any thought, original preserved with history |
| Persistence | None — lost on server restart | SQLite — survives restarts, context resets, new conversations |
| Tagging | Not supported | Semantic tags on any thought |
| Search | Not supported | Search by content or tags |
| Export | Not supported | Markdown, JSON, Mermaid diagrams, ASCII visualization |
| Navigation | Not supported | Retrieve any thought by number, browse filtered history |
Browsing session history
Session data is stored in a standard SQLite database at .maxential/thinking.db. There are several ways to browse past reasoning sessions:
Through the tools themselves — ask AI to use session_list to browse sessions, session_load to restore one, or session_summary for a compressed overview. AI can format, search, and summarize session content conversationally.
With a SQLite browser — open .maxential/thinking.db in any SQLite viewer (VS Code/VSCodium extensions, DB Browser for SQLite, or similar). The schema is straightforward:
| Table | Contains |
|---|---|
sessions |
Session ID, name, description, status, timestamps |
thoughts |
Every thought with its number, content, branch, revision links |
branches |
Branch metadata, status, conclusions, merge history |
tags |
Semantic tags attached to thoughts |
From the terminal:
sqlite3 .maxential/thinking.db "SELECT name, status, datetime(created_at/1000, 'unixepoch', 'localtime') as created FROM sessions ORDER BY updated_at DESC LIMIT 10;"
Installation
Claude Desktop / Claude Code
Add to your MCP configuration:
Claude Desktop config location:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"maxential-thinking": {
"command": "npx",
"args": ["-y", "@bam-devcrew/maxential-thinking-mcp"]
}
}
}
From source
git clone https://github.com/BAM-DevCrew/MAXential-Thinking-MCP.git
cd MAXential-Thinking-MCP
npm install
npm run build
Then configure:
{
"mcpServers": {
"maxential-thinking": {
"command": "node",
"args": ["/path/to/MAXential-Thinking-MCP/dist/src/index.js"]
}
}
}
Configuration
Persistence
Session data is stored in SQLite. By default, the database is created at .maxential/thinking.db in the working directory.
{
"mcpServers": {
"maxential-thinking": {
"command": "npx",
"args": ["-y", "@bam-devcrew/maxential-thinking-mcp"],
"env": {
"MAXENTIAL_DB_PATH": "/path/to/your/thinking.db"
}
}
}
}
MAXENTIAL_DB_PATH value |
Behavior |
|---|---|
| (not set) | .maxential/thinking.db in working directory |
/path/to/file.db |
Use explicit file path |
:memory: |
In-memory only — no persistence across restarts |
If SQLite initialization fails (permissions, native module issues), the server falls back to in-memory mode automatically — it never crashes.
Add .maxential/ to your project's .gitignore to keep session data out of version control.
Logging
{
"mcpServers": {
"maxential-thinking": {
"command": "npx",
"args": ["-y", "@bam-devcrew/maxential-thinking-mcp"],
"env": {
"MAXENTIAL_LOG_FILE": "/path/to/error.log"
}
}
}
}
Usage
You don't call these tools directly — you ask your AI to use MAXential thinking, and it calls the tools as part of its reasoning. Here are examples of what that looks like in practice.
Working through a decision:
use maxential thinking for this - should we use REST or GraphQL for the new API?
AI builds a thought chain analyzing the question, branches to explore each approach separately, adds thoughts with tradeoffs, merges the findings, and reaches a conclusion. The entire reasoning process is numbered, structured, and persisted.
Exploring multiple approaches:
think through the auth redesign using maxential - I want to see branches for JWT, session tokens, and OAuth
AI creates three branches, reasons through each approach independently, then merges the insights back to compare. You can ask it to switch between branches, close dead ends, or dig deeper into a specific path.
Resuming previous thinking:
load up that session where you analyzed our database optimization options
AI browses saved sessions, finds the match, restores it with all thoughts, branches, and tags intact, and continues reasoning from where it left off — even across different conversations.
Reviewing past reasoning:
search your maxential thinking history for anything tagged as a decision
AI searches across the session's tagged thoughts and returns the results. You can also ask it to export the full chain as markdown, or generate a diagram of the thought structure.
Getting a quick summary:
give me a summary of that session - just the key findings, keep it short
AI generates a compressed summary of the session's conclusions, tagged highlights, and branch results — useful for loading context without replaying the full chain.
Development
npm install # Install dependencies
npm run build # Build TypeScript
npm run watch # Watch mode
npm test # Run tests with coverage
npm run test:unit # Unit tests only
npm run test:integration # Integration tests only
License
MIT
Contributing
Issues and PRs welcome at github.com/BAM-DevCrew/MAXential-Thinking-MCP
MAXential Thinking MCP questions
How do I add MAXential Thinking MCP to Claude Code?
Run claude mcp add maxential-thinking-mcp -- npx -y @bam-devcrew/maxential-thinking-mcp in your terminal. Add --scope user to make it available in every project.
How do I add MAXential Thinking MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
What tools does MAXential Thinking MCP provide?
It exposes 20 tools, including think, revise, complete, reset, branch.
How does MAXential handle session persistence?
Sessions are automatically saved to a SQLite database, allowing reasoning to survive server restarts, context window resets, and new conversations.
Can I explore multiple ideas without losing my main thought process?
Yes, you can use the branching tools to create new reasoning paths to explore alternatives and then merge the insights back into the main thread.
What happens if the SQLite database fails to initialize?
The server is designed to be resilient and will automatically fall back to in-memory mode if SQLite initialization fails.
Can I export my reasoning history?
Yes, you can export the thinking chain in either Markdown or JSON formats.