Includes a 'Buy Me a Coffee' link in the README for supporting the developer through donations.
Built with TypeScript, providing type-safe development for the MCP server implementation.
Structured Workflow MCP Server
An MCP server that enforces disciplined programming practices by requiring AI assistants to audit their work and produce verified outputs at each phase of development.
Why I Built This
TLDR: I got tired of repeating "inventory and audit first" across every AI platform and prompt, so I built an MCP server that automatically enforces this disciplined approach. It forces AI to think systematically and follow structured phases instead of jumping straight into code changes.
So I've built an MCP server that fits into my workflow and thinking process while I'm programming. I made it available via npx and you can download it yourself if you want something local.
In essence I was doing some repeated tasks with AI where I wanted it to complete refactoring work for part of a larger project. I was struggling because it was often missing or glossing over key things: classes or systems that already exist (a preferences service for example), creating duplicates of things, or when correcting mistakes, leaving orphaned unused methods/code around places, and when writing tests it would often pull in the wrong imports or put these together in the wrong way resulting in syntax errors but would jump straight into writing the next test without fixing the first one that was broken.
I sort of stumbled on this idea of the model needing to perform an audit and inventory of the current project (or not even the whole project--just one layer or feature in a project) before moving to any kind of implementation phase and it needed a lint iterate lint phase. I tried this with rules with limited success and then prompting with much better success but I was constantly repeating myself.
So I started noodling on this idea of an MCP server that forced the AI to work through a problem in phases or lanes. So that's what this does. There's a number of different workflow styles and I'm open to any other ideas or improvements.
Feel free to check it out if it helps your use case. It's a work in progress but it has been doing a pretty great job for what I'm using it for now. Happy to share more if you are interested.
Features
Enforced Workflow Phases - AI must complete specific phases in order (audit, analysis, planning, implementation, testing, etc.)
Mandatory Output Artifacts - Each phase requires structured documentation or verified outputs before proceeding
Multiple Workflow Types:
- Refactor workflows for code improvement
- Feature development with integrated testing
- Test-focused workflows for coverage improvement
- Test-driven development (TDD) cycles
- Custom workflows for specialized needs
Output Verification - The server validates that outputs contain meaningful content and proper structure
Session State Management - Tracks progress and prevents skipping phases
How It Works
Here's how the AI moves through a structured workflow:
What happens at each step:
- Start Workflow - AI calls a workflow tool (refactor_workflow, create_feature_workflow, etc.)
- AI Gets Phase Guidance - Server provides specific instructions for current phase (audit, analyze, implement, etc.)
- Create Phase Output - AI works through the phase and creates documentation/artifacts
- Auto-Save - Files are automatically saved with numbered naming in task directories
- Phase Validation - Server validates outputs meet requirements before proceeding
- Next Phase - Process repeats until workflow is complete
File Output & Configuration
Automatic File Creation
The server automatically creates numbered workflow files as you progress through phases:
Configuration Options
Default Behavior: Files are saved to ./structured-workflow/[task-name]/
in your current directory.
Custom Output Directory: Configure where files are created by adding CLI parameters to your MCP server config:
Available CLI Parameters:
--output-dir <path>
- Set custom output directory (default:structured-workflow
)--working-dir <path>
- Set working directory for the server--help
- Show help message
Graceful Fallback: If file creation fails (permissions, disk space), the server continues with validation-only mode - your workflow isn't interrupted.
Installation
Quick Start (Recommended) - Zero Installation
Add to your AI assistant config - Uses npx automatically:
💡 Note: I recommend using
@latest
to ensure you always get the newest features and fixes. Without@latest
, npx may cache older versions.
VS Code / Cursor / Windsurf - Add to your MCP settings:
With custom output directory:
Claude Desktop - Add to your claude_desktop_config.json
:
With custom output directory:
Global Installation (Optional)
You can install globally on your machine using NPM:
Then use in your AI assistant config:
With custom output directory:
Auto-Install via Smithery
Smithery provides a number of ways to install directly into your apps including this way for Claude Desktop:
Manual Installation
For developers, you can clone the repository and build it locally:
Usage
Once configured in your AI assistant, start with these workflow tools:
mcp__structured-workflow__build_custom_workflow
- Create custom workflowsmcp__structured-workflow__refactor_workflow
- Structured refactoringmcp__structured-workflow__create_feature_workflow
- Feature developmentmcp__structured-workflow__test_workflow
- Test coverage workflows
Example Output Artifacts
The server enforces that AI produces structured outputs like these:
AUDIT_INVENTORY Phase Output:
COMPARE_ANALYZE Phase Output:
Each phase requires documented analysis and planning before the AI can proceed to implementation.
Tools
Workflow Entry Points
refactor_workflow - Start a structured refactoring process with required analysis and planning phases
create_feature_workflow - Develop new features with integrated testing and documentation requirements
test_workflow - Add test coverage with mandatory analysis of what needs testing
tdd_workflow - Implement Test-Driven Development with enforced Red-Green-Refactor cycles
build_custom_workflow - Create workflows with custom phases and validation requirements
Phase Guidance Tools
- audit_inventory_guidance - Forces thorough code analysis and change cataloging
- compare_analyze_guidance - Requires evaluation of multiple approaches with pros/cons
- question_determine_guidance - Mandates clarification and finalized planning
- phase_output - Validates and records structured outputs from each phase
- workflow_status - Check current progress and validation state
Usage
The server enforces structured workflows through mandatory phases. Each workflow type has different phase requirements:
- Refactor Workflow: AUDIT_INVENTORY → COMPARE_ANALYZE → QUESTION_DETERMINE → WRITE_OR_REFACTOR → LINT → ITERATE → PRESENT
- Feature Workflow: PLANNING → QUESTION_DETERMINE → WRITE_OR_REFACTOR → TEST → LINT → ITERATE → PRESENT
- Test Workflow: AUDIT_INVENTORY → QUESTION_DETERMINE → WRITE_OR_REFACTOR → TEST → ITERATE → PRESENT
- TDD Workflow: PLANNING → WRITE_OR_REFACTOR → TEST → (Red-Green-Refactor cycles) → LINT → PRESENT
Input Validation
The server requires:
task
(string): Description of what you want to accomplishoutputArtifacts
(array): Structured documentation for each completed phase
Output Validation
Each phase completion is validated for:
- Meaningful content length (minimum 10 characters)
- Valid JSON format for structured outputs
- Phase-specific content requirements
- Proper documentation of decisions and analysis
Safety Rule
Files must be read before modification. This prevents accidental data loss and ensures informed changes.
Development
How It Works
- AI starts a workflow using one of the entry point tools
- Server creates a session and tracks phase progression
- Each phase requires specific outputs before proceeding
- The
phase_output
tool validates artifacts have meaningful content - AI cannot skip phases or claim completion without verified outputs
- Session state prevents circumventing the structured approach
Testing the MCP Server
You can quickly try out the Structured Workflow MCP server using the test prompts and helper scripts included in this repository.
- Build the server (if you haven't already):
- Start the server:
- Open the test prompt
docs/test_prompt/mcp_server_test_prompt.md
in your preferred MCP-compatible AI client and paste the contents. - Alternatively, open the sample project located in
refactor-test/
for an end-to-end refactor workflow demo. Follow the steps in itsREADME.md
to run and observe the structured workflow in action. - Watch the AI progress through each phase and verify the structured outputs produced.
Sample Prompts
The docs/sample_prompts
directory contains several ready-to-use prompts illustrating typical workflows:
feature_workflow_prompt.md
refactor_workflow_prompt.md
test_workflow_prompt.md
tdd_workflow_prompt.md
custom_workflow_prompt.md
Use these as a starting point and adapt them to your projects.
Building
The server uses TypeScript with the @modelcontextprotocol/sdk and runs locally via stdio transport.
Pull Requests Welcome
I welcome and encourage pull requests! Whether you're fixing bugs, adding features, or improving documentation, your contributions are valuable.
Please follow these steps:
- Fork the repository on GitHub.
- Create a new branch:
git checkout -b feature/your-feature
. - Make your changes and commit with clear, descriptive messages.
- Write tests for any new functionality and ensure all existing tests pass.
- Push to your branch:
git push origin feature/your-feature
. - Open a pull request and describe your changes clearly.
See CONTRIBUTING.md for more details, if available.
Thank you for contributing!
License
This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License.
This server cannot be installed
local-only server
The server can only run on the client's local machine because it depends on local resources.
Enforces disciplined programming practices by requiring AI assistants to audit their work and produce verified outputs at each phase of development, following structured workflows for refactoring, feature development, and testing.
Related MCP Servers
- -securityAlicense-qualityServes as a guardian of development knowledge, providing AI assistants with curated access to latest documentation and best practices.Last updated -48752TypeScriptMIT License
- -securityAlicense-qualityProvides code manipulation, execution, and version control capabilities. It allows AI assistants to read, write, and execute code while maintaining a history of changes.Last updated -8PythonMIT License
- AsecurityAlicenseAqualityFacilitates interactive feature discussions with AI guidance, maintaining context and providing intelligent recommendations for implementation, architecture, and best practices in software development.Last updated -21JavaScriptMIT License
- -securityAlicense-qualityAI-powered agent that streamlines web testing workflows by allowing developers to record, execute, and discover tests using natural language prompts in their AI coding assistants.Last updated -46PythonApache 2.0