GitHub
GitHub’s official server lets AI read and manage code, issues and pull requests in your account. Use GitHub’s hosted version or run it locally with Docker.
Description from the maker (original English)
Query a codebase map (routes, features, flows, impact) built by static analysis.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"devmap": {
"command": "npx",
"args": [
"-y",
"@flaid/devmap",
"mcp"
]
}
}
}Run this in your terminal
claude mcp add devmap -- npx -y @flaid/devmap mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"devmap": {
"command": "npx",
"args": [
"-y",
"@flaid/devmap",
"mcp"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"devmap": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@flaid/devmap",
"mcp"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Runs on your computer; the tool list appears in your AI app after you install it.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 6, 2026
The registry name was verified with the GitHub account itsflaid, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
This one runs on your computer; there is no internet endpoint
Runs on your computer and uses local settings instead
Runs on your computer; we can only read tool lists from internet endpoints
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).
The GitHub repo last changed –, and the latest version (0.9.0) reached the registry 6 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Node.js on your computer.
In the same category: GitHub, Desktop Commander, Cloudflare MCP
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
GitHub’s official server lets AI read and manage code, issues and pull requests in your account. Use GitHub’s hosted version or run it locally with Docker.
MCP server for terminal commands, file operations, and process management
Cloudflare MCP servers
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Analyze once. Reuse context everywhere. A CLI that maps your codebase for developers and AI agents: entry points, critical files, features and routes.
Documentation · Website · GitHub
Every new AI chat about your code starts from zero — you re-explain the same codebase every time. DevMap fixes that.
It's a CLI that combines static analysis with optional AI interpretation to build a structured, reusable understanding layer of your project — once. That layer (.devmap/index.json, feature maps, DEVMAP.md) is what you and any AI agent read afterward, instead of re-scanning or re-pasting the whole repo. Full repo source stays local — nothing gets sent wholesale to the AI model.
npm install --global @flaid/devmap
Or run without a global install:
npx @flaid/devmap --help
Static analysis still works when AI is not configured.
Run these commands from the root of the project you want to understand:
devmap init
devmap analyze devmap onboarding devmap doctor
devmap init selects a provider, validates its key, stores configuration locally, prepares .devmap/, generates DEVMAP.md, and integrates with AGENTS.md safely.
Choose Groq, OpenRouter, or a custom OpenAI-compatible endpoint with the arrow keys during:
devmap init
Groq keys are available at https://console.groq.com/keys. OpenRouter keys are available at https://openrouter.ai/keys.
For custom endpoints, DevMap asks for the base URL (prefilled with http://localhost:20128/v1) and lists the models your server exposes, so any OpenAI-compatible self-hosted gateway works out of the box.
For OpenRouter, DevMap prompts:
OpenRouter model [openrouter/free]:
Press Enter to use the free router, or type any free or paid OpenRouter model ID. The selected model is saved and used as the primary choice.
For non-interactive Groq setup:
GROQ_API_KEY="your-key" devmap init
On PowerShell:
$env:GROQ_API_KEY="your-key"
devmap init Remove-Item Env:GROQ_API_KEY
For non-interactive OpenRouter setup:
OPENROUTER_API_KEY="your-key" devmap init
The key is stored locally in ~/.devmap/config.json. Requests go directly from your machine to the selected provider. DevMap does not send the key to a DevMap-owned server.
devmap init
devmap analyze devmap analyze --fresh devmap map # full project map devmap map # scoped to one feature or file devmap flow # curated top flows devmap flow devmap explain # file, feature, or function devmap onboarding --write devmap onboarding --write --language id devmap doctor devmap config model auto devmap mcp # run as an MCP server
Groq and OpenRouter use the model selected during init. Change either provider's model with:
devmap config model <model-id>
devmap config model auto
devmap mcp runs DevMap as a Model Context Protocol server over stdio. An AI agent asks the project map questions with tool calls and gets structured JSON back, instead of reading a whole context file up front.
devmap analyze
claude mcp add devmap -- npx -y @flaid/devmap mcp
Other clients launch the same command (npx -y @flaid/devmap mcp) as a stdio server:
{
"mcpServers": { "devmap": { "command": "npx", "args": ["-y", "@flaid/devmap", "mcp"] } } }
Tools: devmap_overview, devmap_map, devmap_flow, devmap_routes, devmap_impact, devmap_find, devmap_analyze and, only with --allow-ai, devmap_explain.
The server is read-only and static by default: no AI provider is contacted unless you start it with --allow-ai, and it only writes inside .devmap/ when the agent calls devmap_analyze or you start it with --auto-refresh. Every answer reports whether the snapshot is stale. See the MCP guide for client setup and tool inputs.
DevMap generates:
.devmap/index.json for lightweight AI-agent navigation.devmap/features/*.json for focused feature maps and source priority.devmap/snapshot.json for the full structured project analysis archiveDEVMAP.md for human and AI-agent usage guidanceAGENTS.md only after confirmation when the file already existsExisting AGENTS.md and DEVMAP.md files are never overwritten.
Read .devmap/index.json first, open the relevant feature map, and inspect its sourcePriority files. Use .devmap/snapshot.json only when the lightweight navigation layer is insufficient.
Use --json for scripts, editors, CI, or AI agents:
devmap analyze --json
devmap onboarding --json devmap doctor --json
JSON mode writes one parseable JSON document to stdout without ANSI colors, terminal decoration, or streamed partial output.
Detection covers the full list above; Next.js and Express receive the deepest analysis. Multi-package workspaces are classified per project type, so a monorepo containing a CLI and a web app is reported as both.
.env files and common generated directories are ignored.MIT