What is detect-harness?

sairaph/detect-harness — explained in plain English

Analysis updated 2026-07-26

★ 1GoAudience · developerComplexity · 3/5Setup · moderate

Mindmap

mindmap
  root((repo))
    What it does
      Detects installed AI assistants
      Generates correct config files
      Preserves existing settings
    Supported assistants
      Claude Desktop and Code
      Cursor and VS Code
      Many others like Zed and Cline
    Tech stack
      Go library and CLI
      Node Python Rust wrappers
      JSON protocol for wrappers
    Safety features
      Rejects invalid configs
      Uses lock files
      Reports partial failures
    Use cases
      Add MCP server support
      Configure multiple assistants
      Support non-Go languages

mindmap root((repo)) What it does Detects installed AI assistants Generates correct config files Preserves existing settings Supported assistants Claude Desktop and Code Cursor and VS Code Many others like Zed and Cline Tech stack Go library and CLI Node Python Rust wrappers JSON protocol for wrappers Safety features Rejects invalid configs Uses lock files Reports partial failures Use cases Add MCP server support Configure multiple assistants Support non-Go languages

What do people build with it?

USE CASE 1

Let users of your dev tool automatically install it as an MCP server in their preferred AI assistant.

USE CASE 2

Generate and apply configuration files across 13 different AI coding assistants from a single definition.

USE CASE 3

Use the library from Node.js, Python, or Rust projects via typed wrappers.

USE CASE 4

Safely update AI assistant configs while preserving existing comments and unrelated settings.

What is it built with?

GoNode.jsPythonRust

How does it compare?

	sairaph/detect-harness	advayc/wrapped	aegrail/aegrail-engine
Stars	1	1	1
Language	Go	Go	Go
Setup difficulty	moderate	easy	hard
Complexity	3/5	2/5	5/5
Audience	developer	general	ops devops

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Requires understanding MCP server definitions and may need building companion binaries for non-Go language wrappers.

So what is it?

detect-harness is a Go library and command-line tool that helps developers add MCP (Model Context Protocol) server installation support to their applications. MCP is a protocol that lets AI coding assistants connect to external tools and services. The problem this project solves is that different AI coding assistants, such as Claude Desktop, Cursor, VS Code, and others, each store their server configurations in different locations and formats. detect-harness handles the work of finding which assistants are installed on a machine and generating the correct configuration for each one. The core idea is that a developer defines their server once, using a simple structure with a name, command, arguments, and environment variables. The library then detects which AI assistants are present, generates a plan showing what configuration changes will be made, and applies those changes safely. The developer's own application controls how selections are presented to the user, while detect-harness handles detection, config generation, conflict handling, and writing files safely. The project supports 13 different AI coding assistants, including Claude Desktop, Claude Code, Cursor, Codex CLI, Gemini CLI, Windsurf, Zed, Cline, Roo Code, Amazon Q, Continue, OpenCode, and VS Code. Each assistant uses a different configuration format: some use plain JSON, others use JSON with comments, TOML, or YAML. The library produces the correct format for each one while preserving existing settings and comments that are unrelated to the server being added. Beyond the Go library, the project provides typed wrappers for Node.js, Python, and Rust. These wrappers communicate with a companion binary through a simple JSON protocol, reading one request and writing one response. This means developers working in those languages can use the same detection and configuration capabilities without rewriting the logic. Safety is a major focus. The library rejects invalid configurations, oversized files, duplicate keys, and symbolic links before writing anything. It uses temporary files with restrictive permissions, rechecks plans immediately before applying them, and uses lock files to prevent concurrent edits. If something goes wrong during a multi-assistant update, partial results are reported rather than hiding failures. The full README is longer than what was shown.

Copy-paste prompts

Prompt 1

Using the detect-harness Go library, write code to define an MCP server named 'my-tool' with a command and arguments, then detect installed AI assistants and print the config plan.

Prompt 2

I have a Node.js app. Show me how to use the detect-harness Node wrapper to install an MCP server config into Cursor and Claude Desktop.

Prompt 3

Write a script using detect-harness that defines a server with environment variables, generates a plan for all detected assistants, and applies the changes safely.

Prompt 4

Explain how to use detect-harness to add MCP server installation support to my existing CLI tool, including how to let the user pick which assistants to configure.

Frequently asked questions