mcp-sequentialthinking-qa

An adaptation of theMCP Sequential Thinking Serverdesigned to guide tool usage in QA and verification processes. Thisserver helps break down verification tasks into manageable steps andprovides LLM-driven recommendations for which MCP tools would be mosteffective at each stage.

A Model Context Protocol (MCP) server that combines sequentialthinking with intelligent tool suggestions for QA tasks. For each stepin the verification process, it provides confidence-scoredrecommendations for which tools to use, along with rationale for whyeach tool would be appropriate.

Features

🤔 Dynamic and reflective QA problem-solving through sequentialthoughts
🔄 Flexible verification process that adapts and evolves
🌳 Support for branching and revision of thoughts
🛠️ LLM-driven intelligent tool recommendations for QA tasks
📊 Confidence scoring for tool suggestions
🔍 Detailed rationale for tool recommendations
📝 Step tracking with expected outcomes
🔄 Progress monitoring with previous and remaining steps
🎯 Alternative tool suggestions for each step
🧠 Memory management with configurable history limits
🗑️ Manual history cleanup capabilities

How It Works

This server facilitates sequential thinking with MCP tool coordinationfor QA and verification tasks. The LLM analyzes available tools andtheir descriptions to make intelligent recommendations forverification processes, which are then tracked and organized by thisserver.

The workflow:

LLM provides available MCP tools to the sequential thinking QAserver
LLM analyzes each verification step and recommends appropriatetools
Server tracks recommendations, maintains context, and managesmemory
LLM executes recommended tools and continues the verificationprocess

Each recommendation includes:

A confidence score (0-1) indicating how well the tool matches theverification need
A clear rationale explaining why the tool would be helpful for thisQA step
A priority level to suggest tool execution order
Suggested input parameters for the tool
Alternative tools that could also be used

The server works with any MCP tools available in your environment andautomatically manages memory to prevent unbounded growth.

Example Usage

Here's an example of how the server guides tool usage for QAverification:

{
	"available_mcp_tools": [
		"mcp-filesystem",
		"mcp-playwright",
		"mcp-omnisearch"
	],
	"thought": "Need to verify package version compatibility before implementing configuration",
	"verification_target": "package version compatibility",
	"current_step": {
		"step_description": "Check installed package version in package.json",
		"expected_outcome": "Confirmed package version and dependencies",
		"recommended_tools": [
			{
				"tool_name": "read_file",
				"confidence": 0.95,
				"rationale": "Examine package.json to determine installed version and dependencies",
				"priority": 1,
				"suggested_inputs": {
					"path": "package.json"
				}
			},
			{
				"tool_name": "execute_command",
				"confidence": 0.75,
				"rationale": "Run npm list to verify actually installed versions",
				"priority": 2,
				"alternatives": ["pnpm list", "yarn list"]
			}
		],
		"next_step_conditions": [
			"Version identified",
			"Dependencies verified",
			"Check for breaking changes between versions"
		]
	},
	"thought_number": 1,
	"total_thoughts": 4,
	"next_thought_needed": true
}

The server tracks your progress and supports:

Creating branches to explore different verification approaches
Revising previous thoughts with new information
Maintaining context across multiple verification steps
Suggesting next steps based on current findings
Adapting to the specific tools available in your environment

Configuration

This server requires configuration through your MCP client. Here areexamples for different environments:

Cline Configuration

Add this to your Cline MCP settings:

{
	"mcpServers": {
		"mcp-sequentialthinking-qa": {
			"command": "npx",
			"args": ["-y", "mcp-sequentialthinking-qa"],
			"env": {
				"MAX_HISTORY_SIZE": "1000"
			}
		}
	}
}

Claude Desktop with WSL Configuration

For WSL environments, add this to your Claude Desktop configuration:

{
	"mcpServers": {
		"mcp-sequentialthinking-qa": {
			"command": "wsl.exe",
			"args": [
				"bash",
				"-c",
				"MAX_HISTORY_SIZE=1000 source ~/.nvm/nvm.sh && /home/username/.nvm/versions/node/v20.12.1/bin/npx mcp-sequentialthinking-qa"
			]
		}
	}
}

API

The server implements a single MCP tool with configurable parameters:

sequentialthinking_qa

A tool for QA-focused sequential thinking with intelligent toolrecommendations for verification tasks.

Parameters:

available_mcp_tools (array, required): Array of MCP tool namesavailable for use (e.g., ["mcp-omnisearch", "mcp-playwright","mcp-filesystem"])
thought (string, required): Your current thinking step in the QAprocess
next_thought_needed (boolean, required): Whether another thoughtstep is needed
thought_number (integer, required): Current thought number
total_thoughts (integer, required): Estimated total thoughtsneeded
verification_target (string, optional): What's being verified(code, config, package version, etc.)
is_revision (boolean, optional): Whether this revises previousthinking
revises_thought (integer, optional): Which thought is beingreconsidered
branch_from_thought (integer, optional): Branching point thoughtnumber
branch_id (string, optional): Branch identifier
needs_more_thoughts (boolean, optional): If more thoughts areneeded
current_step (object, optional): Current step recommendation with:
- step_description: What needs to be done
- recommended_tools: Array of tool recommendations with confidencescores
- expected_outcome: What to expect from this step
- next_step_conditions: Conditions for next step
previous_steps (array, optional): Steps already recommended
remaining_steps (array, optional): High-level descriptions ofupcoming steps

Memory Management

The server includes built-in memory management to prevent unboundedgrowth:

History Limit: Configurable maximum number of thoughts to retain(default: 1000)
Automatic Trimming: History automatically trims when limit isexceeded
Manual Cleanup: Server provides methods to clear history whenneeded

Configuring History Size

You can configure the history size by setting the MAX_HISTORY_SIZEenvironment variable:

{
	"mcpServers": {
		"mcp-sequentialthinking-qa": {
			"command": "npx",
			"args": ["-y", "mcp-sequentialthinking-qa"],
			"env": {
				"MAX_HISTORY_SIZE": "500"
			}
		}
	}
}

Or for local development:

MAX_HISTORY_SIZE=2000 npx mcp-sequentialthinking-qa

Development

Setup

Clone the repository
Install dependencies:

pnpm install

Build the project:

pnpm build

Run in development mode:

pnpm dev

Publishing

The project uses changesets for version management. To publish:

Create a changeset:

pnpm changeset

Version the package:

pnpm changeset version

Publish to npm:

pnpm release

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License - see the LICENSE file for details.

Acknowledgments

Built on theModel Context Protocol
Adapted from theMCP Sequential Thinking Server
Inspired bymcp-sequentialthinking-tools

mcp-sequentialthinking-qa