Agent MCP Server

Git-aware safe file ops for AI agents: delete anything except git-tracked source code. The agent mcp server wraps that behind the Model Context Protocol, so an assistant can use it through 3 defined tools rather than through you.

What it actually does

MCP server (written in Go, single static binary) for git-aware safe file operations. Give your AI agents (Claude Code, GitHub Copilot CLI, Cursor, …) free rein to delete what they build — while making it impossible for them to delete your git-tracked source code.

Its toolset

Everything the assistant can do here goes through one of these:

• safe_delete — Delete files/dirs; refuses git-tracked files or dirs containing them. Supports dry_run
• check_protection — Report whether paths are protected and why
• guard_status — Summarize a directory: tracked (protected) vs untracked (deletable) counts

Configuration

• git on PATH - Go ≥ 1.24 (build only — the binary itself has zero dependencies)

Adding it to your client

Setup follows the usual MCP pattern — install or clone the server, register it in your client's configuration file, restart the client.

When to reach for it

Plenty of developer tooling servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Agent's toolset — safe_delete, check_protection, guard_status — is a fair guide to whether it matches your workflow. It is maintained by Kavachh; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Caveats

• It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
• MCP clients confirm each tool call by default. Leave that on until you have watched what the agent mcp server does with a few real requests.