claude-faf-mcp
Verified Safeby Wolfe-Jam
Overview
Provides AI-optimized context for software projects to large language models, enhancing collaboration and project understanding via 50+ specialized tools within Claude Desktop workflows.
Installation
npx -y claude-faf-mcpEnvironment Variables
- FAF_WORKING_DIR
- MCP_WORKING_DIR
- FAF_MCP_SHOW_ADVANCED
- NO_COLOR
- FAF_DEBUG
Security Notes
The server provides direct local filesystem access (`faf_read`, `faf_write`) and executes `faf-cli` commands via `child_process.exec`. Robust `PathValidator` prevents path traversal attacks and access to sensitive system directories (`/etc`, `/sys`). File size limits are enforced to prevent excessive data exposure. Arguments passed to external CLI commands are explicitly sanitized to mitigate injection risks. An HTTP/SSE server is provided (optional), which could pose a network exposure if not properly secured, but CORS is configurable. No hardcoded secrets were identified. `is_safe_to_run` is true, assuming the user trusts local execution on their machine.
Similar Servers
tmcp
Build Model Context Protocol (MCP) servers for AI agents, providing schema-agnostic tools, resources, and prompts, with optional OAuth 2.1 authentication and distributed session management.
markitdown-mcp
A Model Context Protocol (MCP) server for converting over 29 file formats (PDFs, Office docs, images, audio, etc.) into clean, structured Markdown, primarily for integration with AI workflows and MCP clients like Claude Desktop.
mcp-devtools-server
Standardizes development tool interaction and enhances AI code generation, autocorrection, and workflow automation.
unified-thinking
This MCP server consolidates various cognitive thinking patterns and specialized reasoning tools into a single Go-based implementation for enhanced AI reasoning capabilities within client applications like Claude Desktop.