responsible-vibe-mcp
Verified Safeby mrsimpson
Overview
Transforms any AI coding agent into a structured development partner by guiding it through battle-tested engineering workflows with persistent context and documentation.
Installation
npx responsible-vibe-mcpEnvironment Variables
- PROJECT_PATH
- VIBE_ROLE
- VIBE_WORKFLOW_DOMAINS
- LOG_LEVEL
Security Notes
The server extensively interacts with the local filesystem (reading/writing `.vibe` directory for conversation state, plan files, docs) and executes Git commands via `child_process.execSync`. This is necessary for its core functionality of managing project context and history. While `execSync` carries inherent risks if not used carefully (e.g., with unsanitized user input), the reviewed code snippets use it for controlled internal Git operations. Network communication is handled via the Model Context Protocol SDK, supporting stdio and potentially HTTP/WS, which is expected for an MCP server. The `cli` component also uses `child_process.spawn` for `npm` commands, which is also within the expected operational scope. No obvious patterns of `eval`, code obfuscation, or hardcoded secrets were identified. The primary security risk would arise from the broader agent ecosystem feeding arbitrary, unsanitized input to the server if not properly isolated, rather than from inherent maliciousness or major vulnerabilities in this server's code itself. It is intended to run within a trusted developer environment.
Similar Servers
neurolink
Provides a universal AI development platform with multi-provider support, middleware, tool orchestration, and AI-powered workflow automation.
conductor-tasks
An intelligent assistant for developers to transform requirements into actionable tasks, generate implementation plans, track progress, and accelerate development, leveraging multiple LLMs and integrating into an editor via MCP or as a CLI.
charlie
Charlie is a universal configuration generator that allows developers to define commands, rules, and MCP server configurations once in YAML/Markdown and then generate agent-specific output for various AI agents like Claude, Cursor, and GitHub Copilot.
mcp-server-for-Github
Automates comprehensive GitHub workflows for AI agents, including repository management, issue tracking, pull request operations, code search, and file management, enabling AI-powered development.