LABSCO SUMMARY

Four equations that decide whether an RF number is possible at all.

The useful move is passing in the claim alongside the inputs: the tool returns both the physical limit and a verdict, so "20 MHz at 15 dB SNR gives 500 Mbps" comes back against a Shannon limit of 100.6 Mbps rather than being quietly accepted. The aperture check inside the link budget does the same for antenna gains, which is where dish and patch figures usually go wrong.

What it is

A physics checker for radio engineering. Each tool computes the real figure from the governing equation and, when you supply a claim, says whether the claim is physically possible.

What you get

A full link budget from the Friis transmission equation: free-space path loss, received power and wavelength — with antenna gains rejected when they exceed what the aperture allows
Shannon-Hartley channel capacity for an AWGN channel, spectral efficiency, and any claimed throughput checked against the limit with the exact percentage by which it overshoots
Thermal noise power, which is -174.0 dBm/Hz at the IEEE reference temperature of 290K, cascaded noise figure through multiple stages by the Friis noise formula, system noise temperature and receiver sensitivity
Monostatic radar detection range from the radar range equation, including radar cross-section, integration over multiple pulses and losses, with claimed ranges validated against it
Constants taken from SciPy's CODATA 2018 values, and a warning above 300 GHz where free-space path loss stops describing the atmosphere

Requirements

No account and no key — it is arithmetic on constants. Run it with uvx physbound over stdio; the PyPI package is physbound, currently 0.2.0, and it needs Python 3.12 or newer. The first launch pulls about 60 MB of dependencies including scipy and numpy, so running uvx physbound once in a terminal first makes later starts immediate.

Setup effort

One command — uvx physbound

See installation options

USE CASES

01

Sanity-checking a throughput figure in a proposal

Give the bandwidth and SNR with the claimed rate; the capacity comes back with the overshoot stated as a percentage.

shannon_hartley

02

Working out whether a link will actually close

Compute path loss and received power at the distance and frequency, with the antenna gains checked against their aperture limits.

rf_link_budget

03

Setting the sensitivity target for a receiver

Get the thermal noise floor for the bandwidth, cascade the stage noise figures, and read the resulting sensitivity.

noise_floor

04

Checking a claimed detection range on a small target

Feed peak power, gain, frequency and radar cross-section and compare the maximum range against what was claimed.

radar_range

05

Testing whether stage ordering matters in a receiver chain

Cascade the same stages in both orders and compare the system noise figures that come back.

noise_floor

4 tools

Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.

`noise_floor`

Calculate thermal noise power (kTB), cascaded noise figure, and receiver sensitivity.

Computes the fundamental thermal noise floor N = k_B * T * B, which is -174 dBm/Hz at the IEEE standard temperature of 290K. Optionally cascades multiple amplifier/filter stages using the Friis noise figure formula F_total = F_1 + (F_2-1)/G_1 + (F_3-1)/(G_1*G_2) + ... and computes receiver sensitivity as S_min = N_floor + NF + SNR_required.

Use this tool when you need to:

Determine the thermal noise floor for a receiver bandwidth
Cascade noise figures through a multi-stage receiver chain
Calculate minimum detectable signal / receiver sensitivity
Validate that a claimed noise figure is physically plausible

Returns a PhysicalViolationError dict if inputs violate thermodynamic limits.

Args: bandwidth_hz: Receiver bandwidth in Hz (must be > 0) temperature_k: System noise temperature in Kelvin (default: 290K, must be >= 0) stages: Optional list of stages, each with 'gain_db' and 'noise_figure_db' keys required_snr_db: Required SNR in dB for sensitivity calculation

BEFORE YOU INSTALL

What to know first

Tools you get

4 tools when we connected.

Requires

uvx on PATH

Where it runs

All 3 documented clients run the server on your machine.

CONTEXT COST

Light Standing cost

Rated by the size of the 4 tool definitions that ride along on every turn, not by how many there are

INSTALLATION

Connect your client

Server setup

TERMINAL

uvx physbound

MCP Clients

Claude Desktop, Cursor, Windsurf

JSON

{
  "mcpServers": {
    "physbound": {
      "command": "uvx",
      "args": [
        "physbound"
      ]
    }
  }
}

PhysBound Physics Linter

Four equations that decide whether an RF number is possible at all.

What it is

What you get

Requirements

Setup effort

Sanity-checking a throughput figure in a proposal

Working out whether a link will actually close

Setting the sensitivity target for a receiver

Checking a claimed detection range on a small target

Testing whether stage ordering matters in a receiver chain

4 tools

`noise_floor`

`radar_range`

`rf_link_budget`

`shannon_hartley`

What to know first

Tools you get

Requires

Where it runs

Light Standing cost

Connect your client

Server setup

MCP Clients

Claude Desktop, Cursor, Windsurf

PhysBound Physics Linter

Four equations that decide whether an RF number is possible at all.

What it is

What you get

Requirements

Setup effort

Sanity-checking a throughput figure in a proposal

Working out whether a link will actually close

Setting the sensitivity target for a receiver

Checking a claimed detection range on a small target

Testing whether stage ordering matters in a receiver chain

4 tools

noise_floor

radar_range

rf_link_budget

shannon_hartley

What to know first

Tools you get

Requires

Where it runs

Light Standing cost

Connect your client

Server setup

MCP Clients

Claude Desktop, Cursor, Windsurf

Same category

RFTools Electronics Calculators

PineForge Codegen

ParticlePhysics MCP Server

PsiAnimator-MCP

`noise_floor`

`radar_range`

`rf_link_budget`

`shannon_hartley`