MCP.so
登录
F

Fossil MCP Server

@yfedoseev

基本信息

分类

开发工具

传输方式

stdio

作者

yfedoseev

提交者

Yury Fedoseev

关于 Fossil MCP Server

The code quality toolkit for the vibe coding era.

Fossil MCP Server 配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块,请前往代码仓库查看安装说明。

代码仓库

Fossil MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Fossil Mcp?

Fossil Mcp is a static analysis toolkit purpose-built for projects that use AI-assisted “vibe coding” workflows. It detects dead code, duplicated logic, scaffolding artifacts (like placeholder TODOs and phase comments), and temporary files left behind by AI agents, and it builds a code graph so both developers and AI tools can trace how functions connect. It can run as a standalone CLI, an interactive REPL, a CI/CD check, or a Model Context Protocol (MCP) server that lets AI agents self-check their output.

How to use Fossil Mcp?

Install via the quick install script at fossil‑mcp.com/install.sh (macOS/Linux) or the PowerShell one‑liner for Windows, or download a pre‑built binary from GitHub Releases. Run fossil-mcp with no arguments to scan the current directory and enter the interactive REPL, or use subcommands like fossil-mcp check for CI/CD. To use it as an MCP server, add fossil-mcp (with no arguments) as a command in your AI coding tool’s MCP configuration (e.g., Claude Code, Cursor, VS Code, Windsurf).

Key features of Fossil Mcp

  • Detects unreachable functions, unused exports, and orphaned methods.
  • Finds code clones (exact, renamed, and structural duplicates).
  • Identifies scaffolding artifacts: phase/step comments, TODO/FIXME markers, placeholder bodies.
  • Locates temp and backup files (temp_*, backup_*, old_*, phase_*).
  • Builds a call graph for tracing, blast radius, and code path analysis.
  • Single binary written in Rust, zero configuration, framework‑aware.

Use cases of Fossil Mcp

  • Clean up dead functions left behind after an AI refactoring pass.
  • Discover structurally similar utility functions scattered across modules.
  • Remove stale TODO, “Phase 1”, and placeholder markers accumulated over multiple AI sessions.
  • Audit a codebase for temporary files created by AI tools.
  • Let an AI agent inspect the call graph and blast radius before proposing a refactoring.

常见问题

1

What makes Fossil Mcp different from a traditional linter?

Fossil Mcp is not a general linter — it specifically targets dead code, structural duplication, scaffolding artifacts, and temp files that accumulate when AI writes most of the code and humans review less. It also provides a code graph (not just text search) so agents can trace how functions connect.

2

Does Fossil Mcp require any configuration?

No. It works out of the box with zero configuration. A configuration file is optional if you need custom thresholds or exclusions.

3

How does Fossil Mcp integrate with AI coding tools?

Fossil runs as an MCP server by default. You add `fossil-mcp` as a command in your tool’s MCP configuration (e.g., Claude Code, Cursor, VS Code, Windsurf). Once connected, the AI agent gets tools like `scan_all`, `analyze_dead_code`, `detect_clones`, and `fossil_trace`.

4

What languages does Fossil Mcp support?

Sixteen languages are supported across polyglot codebases. The README mentions framework awareness for React, Next.js, Django, Spring, and Axum, among others.

5

How do I install Fossil Mcp on Windows?

Run the following in PowerShell: `irm fossil-mcp.com/install.ps1 | iex`. Alternatively, download the Windows x86_64 binary from GitHub Releases and add it to your PATH.

评论

开发工具 分类下的更多 MCP 服务器