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

Spec Workflow MCP

Community
A Model Context Protocol server for structured spec-driven development with real-time dashboard and VSCode extension.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #50 of 438
Stars
★ 4.3k Very popular
Transport
stdio (local process)
Runtime
Node.js · Docker
Credentials
No credential needed
License
GPL-3.0
Last commit
65FMRS · C

This server provides comprehensive spec-driven development features, suitable for AI-assisted development projects that need structured workflows and approval mechanisms. Its security features (such as localhost binding, rate limiting, audit logging) make it suitable for enterprise environments, but it lacks built-in HTTPS and user authentication, which must be supplemented via a reverse proxy. The tool list is not explicitly documented, but based on the description, it is inferred to have tools for creating, listing, executing specs, etc.

Strongest · Documentation 16/20 Weakest · Security and permissions 10/20

Reliability
12/20
Security and permissions
10/20
Maintenance
12/20
Documentation
16/20
Setup experience
15/20
Why each score
Reliability 12/20
The README shows a clear project structure with npm package and build commands, but there is no CI configuration or test files to prove the server can actually start and complete the MCP handshake. According to static calibration, reliability cannot exceed 12. Points deducted for lack of verifiable execution evidence.
Security and permissions 10/20
The README describes several security controls (localhost binding, rate limiting, audit logging, security headers, CORS) and provides Docker hardening recommendations. However, these are not verified in the source code, and user authentication/HTTPS are not implemented. There is no evidence of malicious behavior, but permission and confirmation mechanisms are incomplete, so score is capped at 12.
Maintenance 12/20
The repository has 4276 stars, but the README notes that the author is taking a break from maintenance. No license information is provided (not shown in topics). Although the project is popular, there is a lack of maintenance cadence and dependency update records.
Documentation 16/20
The README is comprehensive, covering quick start, client configurations, Docker, security, and documentation links. However, specific tool parameter documentation is missing (there may be a TOOLS-REFERENCE.md but it was not reviewed).
Setup experience 15/20
The README provides configuration examples for multiple clients (Claude, Cursor, etc.) with clear steps. However, it lacks specific environment preparation steps and platform compatibility details, so score is capped at 15.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessReads local filesWrites / deletes local files

Best for

  • Teams that want a structured spec-driven workflow for AI-assisted development.
  • Developers who need to visualize project progress in their IDE.
  • Projects that require an approval process and task tracking for compliance.

Not for

  • Quick prototypes or small projects that don't need a rigid specification process.
  • Scenarios requiring full cloud deployment with user authentication and HTTPS (currently needs a reverse proxy).

Required permissions

  • Read and write access to the project directory (e.g., the `.spec-workflow` folder).
  • Network access: The web dashboard binds to `127.0.0.1:5000` by default, allowing local access only.
  • Create and manage files for specs, tasks, approvals, and logs.

Risks and side effects

  • Data exposure: If the dashboard is exposed to the network, it could be accessed without authorization; it is bound to localhost by default, but reverse proxy and firewall configurations must be cautious.
  • Misuse: Since AI agents may execute commands automatically, the tool may perform unintended actions; it is recommended to use in a controlled environment.
  • Dependency risk: Using npx may be subject to npm supply chain risks; keep the package version up to date.

Setup

Before you start

Runtime：Node.js · Docker

Other optional settings (1)
SPEC_WORKFLOW_HOME optional Redirects global state files to a writable location in sandboxed environments (e.g. /workspace/.spec-workflow-mcp) where $HOME is read-only; not a secret.
  1. Add the following JSON to your MCP client configuration (e.g., Claude Desktop, Cursor, etc.): {"mcpServers":{"spec-workflow":{"command":"npx","args":["-y","@pimzino/spec-workflow-mcp@latest","/path/to/your/project"]}}}.
  2. Choose your interface: A) Start the web dashboard by running npx -y @pimzino/spec-workflow-mcp@latest --dashboard (default port 5000); B) Install the Spec Workflow MCP Extension from the VSCode marketplace.
  3. Replace '/path/to/your/project' with your actual project path. For Claude Code CLI, use claude mcp add spec-workflow npx @pimzino/spec-workflow-mcp@latest -- /path/to/your/project. For sandboxed environments, set the SPEC_WORKFLOW_HOME environment variable to redirect global state files.
claude_desktop_config.json 
{
  "mcpServers": {
    "spec-workflow": {
      "command": "npx",
      "args": [
        "-y",
        "@pimzino/spec-workflow-mcp@latest",
        "/path/to/your/project"
      ]
    }
  }
}

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

.vscode/mcp.json
{
  "servers": {
    "spec-workflow": {
      "command": "npx",
      "args": [
        "-y",
        "@pimzino/spec-workflow-mcp@latest",
        "/path/to/your/project"
      ]
    }
  }
}

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

Terminal
claude mcp add spec-workflow -- npx -y @pimzino/spec-workflow-mcp@latest /path/to/your/project

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

Generate and validate client config

Check that it works

After configuring your MCP client, say "List my specs" to the assistant; a returned (possibly empty) list of specs confirms the server is connected. You can also run npx -y @pimzino/spec-workflow-mcp@latest --dashboard and open http://localhost:5000 to confirm the dashboard works.

Troubleshooting

  1. If the dashboard fails to start, check if port 5000 is occupied or use the `--port` flag to specify another port.
  2. If MCP clients cannot connect, verify the path in the configuration is correct and has read/write permissions.
  3. In sandboxed environments, if you encounter a $HOME read-only error, set `SPEC_WORKFLOW_HOME` to a writable directory.
  4. On Windows, if the `claude mcp add` command fails, try using `cmd.exe /c` approach.

Things to try

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

  • Create a spec for user authentication
  • List my specs
  • Execute task 1.2 in spec user-auth
  • Show the progress of my specs

Use cases

Create specs: Type "Create a spec for user authentication" to generate a complete spec workflow.
Monitor progress: View specs, tasks, and progress in real-time via the web dashboard or VSCode extension.
Approval workflow: Request approval through the dashboard, provide feedback, and track revisions.
Execute tasks: Run specific tasks by saying "Execute task 1.2 in spec user-auth".

Supported clients

Claude Desktop
Claude Code
Augment Code
Cline/Claude Dev
Continue IDE
Cursor
OpenCode
Windsurf
Codex

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

Overview

Spec Workflow MCP is a Model Context Protocol (MCP) server for structured spec-driven development in AI-assisted software development. It provides a sequential workflow for creating specs (Requirements → Design → Tasks), a real-time web dashboard, and a VSCode extension for monitoring and managing project progress directly in your development environment. Features include an approval workflow with revisions, task progress tracking, implementation logs with code statistics, and support for 11 languages.

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 d38e82eaa8a6 Data synced 2026-10-12 Read the FMRS scoring method