⚙️ mcp-devtools: multi-functional development tools MCP server over SSE

by daoch4n

478 downloads

Not rated

GitHub

About

A development tools server for Git management, file operations, AI-assisted editing, and terminal execution, integrable with AI assistants and code editors.

Details

Author: daoch4n
Downloads: 478
Categories: Developer Tools, Other

- Git management: status, stage/commit, diff, log, branch, reset, checkout, show, read file, apply diff
- Direct file operations: search and replace, write to file
- Terminal command execution via execute_command
- AI-assisted file editing using Aider (ai_edit)
- Filesystem access bounded by per-command `

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

Download and install Highlight from highlightai.com/download
Navigate to the plugins tab and select "Add Custom Plugin"
Configure the plugin with the settings below

Plugin Name ⚙️ mcp-devtools: multi-functional development tools MCP server over SSE

Command (node, npx, python, etc.)

Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
Enable "Start Automatically" if you want the plugin to start when Highlight launches

Setting up with Cursor

Open Cursor's settings
Navigate to the MCP section
Click "Add new Global MCP server"
Copy and paste the configuration below based on your operating system

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

Setting up with Claude Desktop

Open Claude Desktop's settings
Go to the Integrations tab
Click "Add MCP Server"
Use the configuration below based on your operating system

Note: Make sure Claude Desktop is running before attempting to connect.

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

Setting up with Windsurf

Open Windsurf's preferences
Go to the Integrations section
Click "Add MCP Integration"
Use the configuration below based on your operating system

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

Setting up with Cline

Open Cline's settings
Navigate to the Integrations section
Click "Add MCP Server"
Copy and paste the configuration below based on your operating system

Note: Make sure to restart Cline after adding the configuration.

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

Setting up with Goose

Open Goose's preferences
Go to the Integrations tab
Select "MCP Servers"
Click "Add New Server"
Use the configuration below based on your operating system

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

From the repository

Install the prerequisite uv (pip install uv), then run uvx mcp-devtools@latest -p 1337 (or clone and run ./server.sh -p 1337 on Linux/macOS, .\server.ps1 -p 1337 on Windows). Configure your MCP client with the URL http://127.0.0.1:1337/sse, set alwaysAllow appropriately, and include the provided system prompt to guide the AI on using repo_path.

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "\u2699\ufe0f mcp-devtools: multi-functional development tools mcp server over sse": {
            "zen-ai-mcp-devtools": {
                "command": "uvx",
                "args": [
                    "mcp-devtools@latest",
                    "-p",
                    "1337"
                ]
            }
        }
    }
}

McpServers

{
    "zen-ai-mcp-devtools": {
        "command": "uvx",
        "args": [
            "mcp-devtools@latest",
            "-p",
            "1337"
        ]
    }
}

⚙️ mcp-devtools: multi-functional development tools MCP server over SSE

- 🔧 mcp-devtools offers a comprehensive suite of development tools:
- 🎋 Git management operations (git_status, git_stage_and_commit, git_diff, git_diff_all, git_log, git_create_branch, git_reset git_checkout, git_show)
- 📁 Git file operations (git_read_file, git_apply_diff)
- 📂 Direct file operations (search_and_replace, write_to_file)
- 🖥️ Terminal commands execution (execute_command)
- 🤖 AI-assisted file operations using Aider (ai_edit (aider_status for systems ready check ))
<br> ℹ️ When using the ai_edit tool please refer to the Aider Configuration documentation.
- 🌐 All these functions are accessible via Server-Sent Events (SSE), making it a powerful and versatile server for various development needs.
- 🛡️ Filesystem access boundaries are maintained via passing repo_path to every file command, so AI assistant only has read/write access to files in the current workspace (or whatever it decides to pass as repo_path , make sure system prompt is solid on that part).
<br> ⚠️ Execise extreme caution with auto-allowing execute_command tool or at least don't leave AI assistant unattended when doing so. MCP server won't stop assistant from execute_command rm -rf ~/

Use Cases

- 🌐 Use it to extend online chat-based assistants such as ChatGPT, Google Gemini or AI Studio, Perplexity, Grok, OpenRouter Chat, DeepSeek, Kagi, T3 Chat with direct access to local files, git, terminal commands execution and AI-assisted file editing capabilities via MCP-SuperAssistant or similar projects.
- 🦘 Use it to boost code editors like Cursor, Windsurf or VSCode extensions like Roo Code, Cline, Copilot or Augment with intuitive Git management and AI-assisted file editing capabilities and say goodbye to those pesky diff application failures wasting your tool calls or Roo having trouble... breaking your carefully engineered automation workflows. Aider seems to get diffs right!
- For Roo Code, place .roomodes into your repo root and Roo will pick it up as 🤖 AI Code mode that 🪃 Orchestrator mode can call.

Prerequisites

pip install uv

Running from pypi

uvx mcp-devtools@latest -p 1337

Running from git

Linux/macOS

git clone "https://github.com/daoch4n/zen-ai-mcp-devtools/" cd zen-ai-mcp-devtools ./server.sh -p 1337

Windows

git clone "https://github.com/daoch4n/zen-ai-mcp-devtools/" cd zen-ai-mcp-devtools .\server.ps1 -p 1337

AI System Prompt

You have development tools at your disposal. Use relevant tools from devtools MCP server for git management, file operations, and terminal access. When using any tool from devtools, always provide the current repository full current working directory path as the 'repo_path' option, do not set it to any other folder. 'repo_path' must be explicitly asked from user in beginning of conversation. When using execute_command tool, the current working directory will be set to repo_path provided. When using it for file manipulations, make sure to pass full path in the terminal command including repo_path prefix as manipulated file path.

MCP Server Configuration

To integrate mcp-devtools with your AI assistant, add the following configuration to your MCP settings file:

{ "mcpServers": { "devtools": { "url": "http://127.0.0.1:1337/sse", "disabled": false, "alwaysAllow": [], "timeout": 300 } } }

Known Issues and Workarounds

Issue:

write_to_file and 💾 Direct Code Editing vs 🤖 Delegated Editing by Coding Agent

🔍 When using the write_to_file tool for direct code editing, especially with languages like JavaScript that utilize template literals (strings enclosed by backticks), you may encounter unexpected syntax errors. This issue stems from how the AI assistant generates the content string, where backticks and dollar signs within template literals might be incorrectly escaped with extra backslashes (\).

Mitigation:

🔨 The write_to_file tool integrates with tsc (TypeScript compiler) for .js, .mjs, and .ts files. The output of tsc --noEmit --allowJs is provided as part of the tool's response. AI assistants should parse this output to detect any compiler errors and should not proceed with further actions if errors are reported, indicating a problem with the written code.

Workarounds:

🤖 (most reliable) Instruct your AI assistant to delegate editing files to MCP-compatible coding agent by adding it as another MCP server, as it is more suitable for direct code manipulation, and let AI assistant act as task orchestrator that will write down plans and docs with write_to_file and delegate coding to specialized agent, then use git_read_file or git_diff to check up on agent's work, and manage commits and branches Aider via its MCP bridge is already integrated as ai_edit tool).

🖥️ (if you're feeling lucky) Instruct your AI assistant to craft a terminal command to edit problematic file via execute_command tool.

Available Tools

`git_status`

- Description: Shows the working tree status. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      }
    },
    "required": [
      "repo_path"
    ]
  }

`git_diff_all`

- Description: Shows all changes in the working directory (staged and unstaged, compared to HEAD). - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      }
    },
    "required": [
      "repo_path"
    ]
  }

`git_diff`

- Description: Shows differences between branches or commits. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "target": {
        "type": "string"
      }
    },
    "required": [
      "repo_path",
      "target"
    ]
  }

`git_stage_and_commit`

- Description: Records changes to the repository. If files are provided, only those files will be staged and committed. If files are not provided, all changes in the working directory will be staged and committed. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "message": {
        "type": "string"
      },
      "files": {
        "type": "array",
        "items": {
          "type": "string"
        },
        "nullable": true
      }
    },
    "required": [
      "repo_path",
      "message"
    ]
  }

`git_reset`

- Description: Unstages all staged changes. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      }
    },
    "required": [
      "repo_path"
    ]
  }

`git_log`

- Description: Shows the commit logs. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "max_count": {
        "type": "integer",
        "default": 10
      }
    },
    "required": [
      "repo_path"
    ]
  }

`git_create_branch`

- Description: Creates a new branch from an optional base branch. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "branch_name": {
        "type": "string"
      },
      "base_branch": {
        "type": "string",
        "nullable": true
      }
    },
    "required": [
      "repo_path",
      "branch_name"
    ]
  }

`git_checkout`

- Description: Switches branches. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "branch_name": {
        "type": "string"
      }
    },
    "required": [
      "repo_path",
      "branch_name"
    ]
  }

`git_show`

- Description: Shows the contents of a commit. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "revision": {
        "type": "string"
      }
    },
    "required": [
      "repo_path",
      "revision"
    ]
  }

`git_apply_diff`

- Description: Applies a diff to the working directory. Also outputs a diff of the changes made after successful application and tsc --noEmit --allowJs output for .js, .mjs, and .ts files to facilitate clean edits. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "diff_content": {
        "type": "string"
      }
    },
    "required": [
      "repo_path",
      "diff_content"
    ]
  }

`git_read_file`

- Description: Reads the content of a file in the repository. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "file_path": {
        "type": "string"
      }
    },
    "required": [
      "repo_path",
      "file_path"
    ]
  }

`search_and_replace`

- Description: Searches for a string or regex pattern in a file and replaces it with another string. It first attempts to use sed for the replacement. If sed fails or makes no changes, it falls back to a Python-based logic that first attempts a literal search and then a regex search if no literal matches are found. Also outputs a diff of the changes made after successful replacement and tsc --noEmit --allowJs output for .js, .mjs, and .ts files to facilitate clean edits. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "file_path": {
        "type": "string"
      },
      "search_string": {
        "type": "string"
      },
      "replace_string": {
        "type": "string"
      },
      "ignore_case": {
        "type": "boolean",
        "default": false
      },
      "start_line": {
        "type": "integer",
        "nullable": true
      },
      "end_line": {
        "type": "integer",
        "nullable": true
      }
    },
    "required": [
      "repo_path",
      "file_path",
      "search_string",
      "replace_string"
    ]
  }

`write_to_file`

- Description: Writes content to a specified file, creating it if it doesn't exist or overwriting it if it does. Also outputs a diff of the changes made after successful write and tsc --noemit --allowJs output for .js .mjs .ts files to facilitate clean edits. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "file_path": {
        "type": "string"
      },
      "content": {
        "type": "string"
      }
    },
    "required": [
      "repo_path",
      "file_path",
      "content"
    ]
  }

`execute_command`

- Description: Executes a custom shell command. The repo_path parameter is used to set the current working directory (cwd) for the executed command. - Input Schema:

  {
    "type": "object",
    "properties": {
      "repo_path": {
        "type": "string"
      },
      "command": {
        "type": "string"
      }
    },
    "required": [
      "repo_path",
      "command"
    ]
  }

`ai_edit`

- Description: AI pair programming tool for making targeted code changes using Aider. Use this tool to: 1. Implement new features or functionality in existing code 2. Add tests to an existing codebase 3. Fix bugs in code 4. Refactor or improve existing code 5. Make structural changes across multiple files

The tool requires:
- A repository path where the code exists
- A detailed message describing what changes to make. Please only describe one change per message. If you need to make multiple changes, please submit multiple requests.

Best practices for messages:
- Be specific about what files or components to modify
- Describe the desired behavior or functionality clearly
- Provide context about the existing codebase structure
- Include any constraints or requirements to follow

Examples of good messages:
- "Add unit tests for the Customer class in src/models/customer.rb testing the validation logic"
- "Implement pagination for the user listing API in the controllers/users_controller.js file"
- "Fix the bug in utils/date_formatter.py where dates before 1970 aren't handled correctly"
- "Refactor the authentication middleware in middleware/auth.js to use async/await instead of callbacks"
- Input Schema:

  {

    "type": "object",

    "properties": {

      "repo_path": {

        "type": "string"

      },

      "message": {

        "type": "string"

      },

      "options": {

        "type": "array",

        "items": {

          "type": "string"

        },

        "nullable": true

      }

    },

    "required": [

      "repo_path",

      "message"

    ]

  }

`aider_status`

- Description: Check the status of Aider and its environment. Use this to: 1. Verify Aider is correctly installed 2. Check that API keys are set up 3. View the current configuration 4. Diagnose connection or setup issues - Input Schema: ```json { "type": "object", "properties": { "repo_path": { "type": "string" }, "check_environment": { "type": "boolean", "default": true } }, "required": [ "repo_path" ] }

No reviews yet — be the first

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.

About

Details

Jump to

Setting up with Highlight

From the repository

Claude Desktop / Cursor

McpServers

⚙️ mcp-devtools: multi-functional development tools MCP server over SSE

Use Cases

Prerequisites

Running from pypi

Running from git

Linux/macOS

Windows

AI System Prompt

MCP Server Configuration

Known Issues and Workarounds

write_to_file and 💾 Direct Code Editing vs 🤖 Delegated Editing by Coding Agent

Available Tools

git_status

git_diff_all

git_diff

git_stage_and_commit

git_reset

git_log

git_create_branch

git_checkout

git_show

git_apply_diff

git_read_file

search_and_replace

write_to_file

execute_command

ai_edit

aider_status

Install

More like this

Sequential Thinking

Sentry

Github

Filesystem

EverArt

Google Drive

`write_to_file` and 💾 Direct Code Editing vs 🤖 Delegated Editing by Coding Agent

`git_status`

`git_diff_all`

`git_diff`

`git_stage_and_commit`

`git_reset`

`git_log`

`git_create_branch`

`git_checkout`

`git_show`

`git_apply_diff`

`git_read_file`

`search_and_replace`

`write_to_file`

`execute_command`

`ai_edit`

`aider_status`