Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
N1MM Logger+ MCP server. Live contest state: QSOs, bandmap, score, rate, multipliers.
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": {
"n1mm": {
"command": "uvx",
"args": [
"n1mm-mcp"
]
}
}
}Run this in your terminal
claude mcp add n1mm -- uvx n1mm-mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"n1mm": {
"command": "uvx",
"args": [
"n1mm-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": {
"n1mm": {
"type": "stdio",
"command": "uvx",
"args": [
"n1mm-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 8, 2026
The registry name was verified with the GitHub account qso-graph, 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 (0.4.0) reached the registry 12 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Python (uv) on your computer.
In the same category: Spec Driven Development, Remote Desktop Commander, Winecellarbuilders
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
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Find the team for your wine room.
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.
MCP server for N1MM Logger+: live contest state — station, QSOs, bandmap, score and rate, multipliers, and contest clock — through any MCP-compatible AI assistant.
Data from N1MM Logger+'s UDP broadcasts on your local network. Part of the qso-graph project. No authentication required.
uvx n1mm-mcp # run it; nothing to install
| Tool | Description | Key Parameters |
|---|---|---|
n1mm_current_state | Station snapshot: connection, contest, operator, radios | station_name |
n1mm_lookup | The callsign being entered (pre-log) plus current band and mode | station_name |
n1mm_contacts | QSO log: recent contacts, edits and deletes | count, since, band, mode, call_pattern |
n1mm_bandmap | Live spots, multiplier targets, band activity | band, mode, callsign, mults_only |
n1mm_performance | Score, rate, bands, run/S&P, hourly timeline | band, mode |
n1mm_multipliers | Multiplier grid, needs, value analysis | band |
n1mm_clock | Contest timing, off-time, pacing | duration_hours, target_score, target_qsos, min_gap_minutes |
n1mm_diagnostics | Server health, parse errors, memory | station_name |
n1mm_so2r | Two radios: each radio now, transmit/receive focus, same-band warning, transmit time and minutes per band, focus swaps, QSOs per radio and while the other radio runs | station_name |
n1mm_network | Every networked station: operator, radios, rate, idle time, band, shared score per contest call, stations sharing a band and mode, QSOs per operator | — |
n1mm_pileup | DXpedition pileup: rate, unique calls, dupes, continents and countries calling, split, dead air | window_minutes |
get_version_info | Service version + upstream spec version (fleet identity attestation) | — |
Every tool except n1mm_network (which covers all stations) takes an optional station_name for multi-station setups.
Everything comes from N1MM's UDP broadcasts; n1mm-mcp never talks to a radio. N1MM reports transmitting only when it keys the radio itself (macros, its CW keyer), so a paddle or microphone PTT doesn't count as transmit time in n1mm_so2r.
N1MM Logger+ is a Windows contest logger. It can broadcast its state over UDP: contacts, spots, radio info and score. n1mm-mcp listens to those broadcasts and keeps the contest state in memory, so an assistant can answer questions about it. N1MM doesn't know it's there.
N1MM Logger+ (Windows)
│ UDP broadcast (port 12060, XML) ▼ n1mm-mcp (Python: the same PC, or any OS on the LAN) ├── UDP listener (background) ├── State engine (in memory, per StationName) │ MCP protocol (stdio) ▼ AI assistant
127.0.0.1:12060, N1MM's address for "this PC". n1mm-mcp listens there by default, and nothing outside the PC can reach it.Give one address of this PC and name who may send:
n1mm-mcp --bind 192.168.1.20 --allow 192.168.1.40
Then point N1MM at this PC's address (192.168.1.20:12060), or at your subnet's broadcast address (192.168.1.255:12060) to reach every PC on it. N1MM accepts several destinations separated by spaces. Don't use 255.255.255.255: N1MM's own documentation warns that it risks broadcasting to the internet (N1MM: External UDP Broadcasts).
--allow is required whenever --bind is a network address, and is repeatable for more than one PC. A datagram from any other address is dropped without being read, and counted — n1mm_diagnostics shows the count and the addresses, so traffic you did not expect is visible rather than silent.
`--bind 0.0.0.0` is refused. It listened on every interface at once, including whatever Wi-Fi a laptop happened to be joined to, and it cannot be narrowed afterwards. Name the interface you mean.
A UDP sender's address can be forged by anything on the same network, so the list keeps other machines out; it is not proof of who sent a datagram. What limits the damage is that n1mm-mcp only reads: it has no tool that writes to N1MM, your log, or anything else. A forged packet can therefore mislead you about contest state — a wrong frequency, a spot that was never spotted — but it cannot act, and nothing it contains is written anywhere. Treat the state as reported, not as verified, and run on a network you trust.
n1mm-mcp works with any MCP-compatible client. Add the server config and restart. The tools appear automatically.
Add to claude_desktop_config.json (~/Library/Application Support/Claude/ on macOS, %APPDATA%\Claude\ on Windows):
{
"mcpServers": { "n1mm": { "command": "uvx", "args": ["n1mm-mcp"] } } }
Add to .claude/settings.json:
{
"mcpServers": { "n1mm": { "command": "uvx", "args": ["n1mm-mcp"] } } }
{
"mcpServers": { "n1mm": { "command": "uvx", "args": ["n1mm-mcp"] } } }
Add to .cursor/mcp.json (project-level) or ~/.cursor/mcp.json (global):
{
"mcpServers": { "n1mm": { "command": "uvx", "args": ["n1mm-mcp"] } } }
Add to .vscode/mcp.json in your workspace:
{
"servers": { "n1mm": { "command": "uvx", "args": ["n1mm-mcp"] } } }
Add to ~/.gemini/settings.json (global) or .gemini/settings.json (project):
{
"mcpServers": { "n1mm": { "command": "uvx", "args": ["n1mm-mcp"] } } }
"What's my rate over the last hour?"
"Which multipliers do I still need on 20m?"
"Is the station on the bandmap a new multiplier?"
"How much off-time have I used, and am I on pace for my target?"
"Show me the last 10 QSOs."
| Option | Default | Description |
|---|---|---|
--port | 12060 | UDP listen port |
--bind | 127.0.0.1 | Address to listen on. One address of this PC; 0.0.0.0 is refused |
--allow | this PC | An address that may send to us; repeatable. Required with a network --bind |
--transport | stdio | MCP transport (stdio or streamable-http) |
--heartbeat-timeout | 60 | Seconds before the connection goes stale |
--stale-timeout | 900 | Seconds before the connection goes disconnected |
--max-spots | 2000 | Maximum spots in the bandmap buffer |
--spot-ttl | 30 | Spot time-to-live in minutes |
N1MM_MCP_MOCK=1 n1mm-mcp
n1mm-mcp --transport streamable-http
Then open the MCP Inspector at http://localhost:8008.
git clone https://github.com/qso-graph/n1mm-mcp.git
cd n1mm-mcp uv sync --group dev uv run pytest
GPL-3.0-or-later