physbound
PhysBound is an MCP server that validates RF and physics calculations against hard physical limits, catching AI hallucinations in engineering workflows.
RF link budgets: Compute FSPL and received power via the Friis equation, and validate antenna gains against aperture/Harrington limits.
Shannon-Hartley capacity: Calculate the maximum channel capacity and validate whether a claimed throughput is physically possible.
Noise floor and sensitivity: Compute thermal noise (kTB), cascade noise figures with the Friis formula, and determine receiver sensitivity.
Radar range: Calculate maximum monostatic detection range and validate range claims, including the fourth-root power/range relationship.
Antenna analysis: Determine gain limits, beamwidth, far-field distance, and implied aperture efficiency for a single antenna.
Radar ambiguity: Compute unambiguous range and velocity, Doppler aliasing, and the range-Doppler dilemma invariant.
Violations return structured PhysicalViolationError responses with the violated law, computed limit, and LaTeX explanation.
Generates structured physical violation errors with LaTeX-formatted explanations to provide clear, mathematically rigorous reasoning for rejected calculations.
LLMs generate plausible-sounding RF numbers that violate fundamental physics — throughput above the Shannon limit, antenna gains no aperture can produce, radar ranges the range equation forbids. PhysBound gives any MCP-compatible AI assistant six validated calculation tools, backed by CODATA constants (via SciPy) and dimensional analysis (via Pint). Impossible claims return structured PhysicalViolationError responses with the violated law, the computed limit, and a LaTeX explanation — not silent failures.
Tool | What it validates |
Friis link budgets: FSPL, received power, antenna gains vs. aperture/Harrington limits | |
Throughput claims against channel capacity | |
Thermal noise | |
Detection-range claims against the monostatic radar range equation | |
Gain limits, beamwidth, and far-field distance for a single antenna | |
Pulse-Doppler unambiguous range/velocity, Doppler aliasing, range resolution |
Installation
PhysBound is a standard stdio MCP server published on PyPI. The recommended launch command is uvx physbound, which fetches and runs the latest release in an isolated environment — no manual install step.
First run:
uvxdownloads ~60 MB of dependencies (SciPy, NumPy) the first time. Runuvx physboundonce in a terminal to pre-cache them (Ctrl-C to exit); subsequent starts are instant.
Claude Code
claude mcp add physbound -- uvx physboundClaude Desktop
Add to claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\):
{
"mcpServers": {
"physbound": {
"command": "uvx",
"args": ["physbound"]
}
}
}Cursor, Windsurf, and other MCP clients
Use the same JSON server entry as above in your client's MCP configuration file (Cursor: ~/.cursor/mcp.json; Windsurf: ~/.codeium/windsurf/mcp_config.json).
Without uv
If you prefer a plain Python install (requires Python 3.12+):
pip install physboundthen set "command": "physbound" (with no args) in the client configuration.
Once configured, ask your assistant an RF question — "Can a 20 MHz channel with 15 dB SNR support 500 Mbps?" — and it will answer with physics-validated numbers.
Related MCP server: physics-lint-mcp
Command-Line Usage
The same validators are available directly from the terminal — useful in scripts and CI pipelines, since a physics violation exits with code 1:
$ physbound check shannon --bandwidth-hz 20e6 --snr-db 15 --claimed-throughput-bps 500e6
PHYSICS VIOLATION [Shannon-Hartley Theorem]
Claimed throughput 500000000.0 bps exceeds Shannon limit of 100556153.5 bps by 397.2%
Computed limit: 1.00556e+08 bps
Claimed value: 5e+08 bps
$ echo $?
1$ physbound check shannon --bandwidth-hz 20e6 --snr-db 15 --claimed-throughput-bps 100e6
Shannon-Hartley Capacity: ... # exit code 0Available subcommands: check {link-budget, shannon, noise, radar-range, antenna, radar-ambiguity} (flags mirror the MCP tool parameters; add --json for the full structured result) and serve --transport {stdio,http} for running the MCP server over HTTP. See physbound check --help for details.
What LLMs Get Wrong
Sixteen real hallucination patterns, each caught by PhysBound's test suite:
# | Category | LLM Hallucination | PhysBound Truth | Verdict |
1 | Shannon-Hartley | A 20 MHz 802.11n channel with 15 dB SNR can achieve 500 Mbps | Shannon limit: 100.6 Mbps (not 500 Mbps) | CAUGHT |
2 | Shannon-Hartley | A 100 MHz 5G channel with 20 dB SNR delivers 2 Gbps | Shannon limit: 665.8 Mbps (not 2000 Mbps) | CAUGHT |
3 | Antenna Aperture | A 30 cm dish antenna at 1 GHz provides 45 dBi gain | Physical limit: 12.1 dBi (Harrington); aperture eta=1: 9.9 dBi; typical dish: 7.4 dBi (eta=0.55) (not 45 dBi) | CAUGHT |
4 | Thermal Noise | Receiver noise floor of -180 dBm/Hz at room temperature | Thermal noise floor: -174.0 dBm/Hz at 290K (not -180 dBm/Hz) | CAUGHT |
5 | Link Budget / FSPL | Wi-Fi at 2.4 GHz with 20 dBm TX reaches 10 km with -40 dBm RX power | Actual RX power at 10 km: -94.1 dBm (not -40 dBm) | CAUGHT |
6 | Link Budget / FSPL | A 1W transmitter at 12 GHz with 0 dBi antennas reaches GEO at -80 dBm | Actual RX power at GEO: -175.1 dBm (not -80 dBm) | CAUGHT |
7 | Link Budget / FSPL | Bluetooth at 2.4 GHz with 0 dBm TX and 0 dBi antennas reaches 1 km at -60 dBm | Actual RX power at 1 km: -100.1 dBm (not -60 dBm) | CAUGHT |
8 | Shannon-Hartley | A 10 MHz LTE channel at 10 dB SNR supports 1 Gbps | Shannon limit: 34.6 Mbps (not 1000 Mbps) | CAUGHT |
9 | Noise Cascade | Receiver NF is the same regardless of stage order: LNA(20dB/1.5dB) + Mixer(10dB/8dB) | LNA first: 1.66 dB vs mixer first: 8.03 dB (penalty: 6.4 dB) | CAUGHT |
10 | Antenna Aperture | A 10 cm patch antenna at 900 MHz provides 20 dBi gain | Physical limit: 4.4 dBi (Harrington, D < lambda); aperture eta=1: -0.5 dBi; typical: -3.1 dBi (eta=0.55) (not 20 dBi) | CAUGHT |
11 | Radar Range Equation | Doubling transmit power doubles radar detection range | Range increases by factor 1.189 (2^(1/4) = 1.189), not 2.0 | CAUGHT |
12 | Radar Range Equation | Small drone (0.01 m^2 RCS) detectable at 200 km by 1 kW X-band radar with 30 dBi gain | Max range: 2.7 km for 0.01 m^2 RCS at 1 kW X-band (not 200 km) | CAUGHT |
13 | Antenna Gain | A 0.5 m user-terminal dish at 12 GHz gives 50 dBi gain | Physical limit: 36.1 dBi (Harrington); typical dish: 33.4 dBi (eta=0.55) (not 50 dBi) | CAUGHT |
14 | Antenna Gain | A 3 m dish at 10 GHz is in its far field at 10 m, so gain can be measured there | Far-field distance 2D^2/lambda = 600 m; 10 m is in the near field | CAUGHT |
15 | Radar Ambiguity | A 10 GHz radar at 10 kHz PRF unambiguously measures 500 m/s | v_ua = lambda*PRF/4 = +/-74.9 m/s (blind speed 149.9 m/s); 500 m/s aliases (not 500 m/s) | CAUGHT |
16 | Radar Ambiguity | A 10 GHz pulse-Doppler radar can unambiguously cover 150 km and +/-300 m/s at once | R_uav_ua = clambda/8 = 1.123e+06 m^2/s for any PRF; claimed 4.5e+07 m^2/s (40x too large) | CAUGHT |
Generated automatically by pytest tests/test_marketing.py -s
Tools
Full derivations, sources, and worked examples for every formula are in docs/formulas.md.
rf_link_budget
Computes a complete RF link budget using the Friis transmission equation and validates antenna gains against physical limits.
Example: "What's the received power for a 2.4 GHz link at 100 m with 20 dBm TX, 10 dBi TX gain, 3 dBi RX gain?"
Returns FSPL, received power, wavelength, and — when antenna diameters are supplied — gain limit checks. Gains above the hard bound G_max = max((pi D / lambda)^2, (ka)^2 + 2ka) (the eta = 1 aperture value, or Harrington's bound for electrically small antennas) are rejected; gains above the typical dish value (eta = 0.55) are accepted with a warning. Also warns inside the far-field distance 2D^2/lambda and rejects negative losses.
shannon_hartley
Computes Shannon-Hartley channel capacity C = B log2(1 + SNR) and validates throughput claims.
Example: "Can a 20 MHz channel with 15 dB SNR support 500 Mbps?"
Returns theoretical capacity, spectral efficiency, and whether the claim is physically possible, including the exact percentage by which a violating claim exceeds the Shannon limit.
noise_floor
Computes thermal noise power N = k_B T B, cascades noise figures through multi-stage receivers with the Friis noise formula, and calculates receiver sensitivity.
Example: "What's the noise floor for a 1 MHz receiver at 290 K with a two-stage LNA chain?"
Returns thermal noise in dBm and watts, cascaded noise figure, effective input noise temperature T_e = 290 K * (F - 1), and receiver sensitivity.
radar_range
Computes the monostatic radar range equation R_max = [P_t G^2 lambda^2 sigma / ((4pi)^3 S_min L)]^(1/4) and validates detection-range claims.
Example: "Can a 1 kW X-band radar with 30 dBi gain detect a 0.01 m^2 drone at 200 km?"
Returns maximum detection range, minimum detectable signal, wavelength, and intermediate values. Catches the common fourth-root fallacy that doubling power doubles range.
antenna_gain
Analyses a single antenna from its diameter or physical area: gain limits, beamwidth, and far-field distance, with optional validation of a claimed gain.
Example: "Can a 0.5 m dish at 12 GHz really give 50 dBi? What beamwidth should I expect?"
Returns the physical gain limit (aperture or Harrington bound, and which one governs), typical gain at the given efficiency (default 0.55), effective aperture, half-power beamwidth estimates, far-field distance 2D^2/lambda, and — for a claimed gain — the implied aperture efficiency and validity.
radar_ambiguity
Computes pulse-Doppler ambiguity limits for a given carrier frequency and PRF.
Example: "Can a 10 GHz radar at 10 kHz PRF unambiguously measure a 500 m/s target out to 150 km?"
Returns unambiguous range R_ua = c / (2 PRF), unambiguous velocity v_ua = lambda PRF / 4, blind speed, Doppler shift and apparent (aliased) velocity, range resolution (c tau / 2, or c / 2B with pulse compression), minimum range, duty cycle, and the range-Doppler dilemma invariant R_ua v_ua = c lambda / 8. Rejects range, velocity, or resolution claims the PRF or pulse cannot support.
Physics Guarantees
Every calculation is validated against hard physical limits:
Speed of light:
c = 299,792,458 m/s— no exceptionsThermal noise floor:
N = -174 dBm/Hzat 290 K — the IEEE standard referenceShannon limit:
C = B log2(1 + SNR)— no throughput claim exceeds thisAntenna gain limit:
G_max = max((pi D / lambda)^2, (ka)^2 + 2ka)— the eta = 1 aperture value for planar apertures and Harrington's bound for any antenna in a sphere of diameter D; eta = 0.55 is a warning threshold, not a limitRadar range equation:
R_max = [P_t G^2 lambda^2 sigma / ((4pi)^3 S_min)]^(1/4)— range obeys the fourth-root lawReceiver sensitivity:
S_min = k (T_A + T_0 (F - 1)) B * SNRwithT_e = T_0 (F - 1)referenced to 290 KUnambiguous range:
R_ua = c / (2 PRF)— echoes beyond it fold into a later pulse intervalUnambiguous velocity:
v_ua = lambda PRF / 4— Doppler is sampled at the PRF, so|f_d| <= PRF/2Range-Doppler dilemma:
R_ua * v_ua = c lambda / 8— no PRF choice beats it for a given carrier
Violations return structured PhysicalViolationError responses with LaTeX explanations, not silent failures.
Examples and Documentation
Catching Hallucinations — walkthrough of five real LLM failure modes with full JSON responses
Interactive Demo Notebook — Jupyter notebook calling the physics engines directly
Formula Reference — every formula, constant, and validation guard with sources
Changelog — release history
Development
Requires uv and Python 3.12+ (CI covers 3.12–3.14):
# Clone and install
git clone https://github.com/JonesRobM/physbound.git
cd physbound
uv sync --all-extras
# Run checks
uv run pytest tests/ -v
uv run ruff check src/ tests/
uv run mypy src/physbound/
# Regenerate the hallucination table above
uv run pytest tests/test_marketing.py -s
# Start the MCP server locally
uv run physboundContributions that expand the set of validated physics domains are welcome — see CONTRIBUTING.md for the architecture guide and the step-by-step recipe for adding a new validator.
Use Cases
RF system design review — validate link budgets, receiver sensitivity, and noise cascades
Telecom proposal vetting — catch impossible throughput claims before they reach a customer
Radar system sizing — sanity-check detection range, PRF selection, and ambiguity trade-offs
Educational tools — teach Shannon-Hartley, Friis transmission, and thermal noise with verified calculations
CI/CD for physics — integrate as a validation step in engineering pipelines
Support
If PhysBound is useful in your work, consider buying me a coffee.
License
MIT License. See LICENSE.
Related
Model Context Protocol — the open standard for AI tool integration
MCP Server Registry — official directory of MCP servers
FastMCP — Python framework for building MCP servers
Available Tools
6 toolsantenna_gainA
Calculate antenna aperture gain limits, beamwidth, far-field distance; validate gain claims.
For a circular aperture of diameter D (or an aperture of area A, converted to the equivalent circular diameter D = sqrt(4A/pi)) computes the planar-aperture directivity D_0 = 4piA/lambda^2 = (piD/lambda)^2 (aperture efficiency eta = 1), Harrington's bound D_max = (ka)^2 + 2ka (k = 2pi/lambda, a = D/2) for any antenna enclosed in a sphere of diameter D, the hard physical gain limit max(D_0, D_max) (numerically D_max; equal to D_0 to within 0.1 dB for D >> lambda, but several dB higher for electrically small antennas D < lambda, e.g. a 2.15 dBi dipole in a 0.1 m footprint at 900 MHz is valid although D_0 = -0.5 dBi), and the typical gain G = etaD_0 at the given efficiency (default eta = 0.55, parabolic dish). The limiting_bound field reports 'harrington' for D < lambda and 'aperture' for D >= lambda. Also returns the effective aperture A_e = etaA, a half-power beamwidth estimate (HPBW ~ 70lambda/D degrees for a tapered reflector; 58.4lambda/D for a uniformly illuminated circular aperture) and the far-field (Fraunhofer) distance 2*D^2/lambda.
If claimed_gain_dbi is given it is validated: a claim above the physical limit is rejected as a physics violation; a claim between the typical gain and the physical limit is accepted with a warning and the implied planar-aperture efficiency eta = G_claim / D_0 is reported.
Use this tool when you need to:
Check whether a quoted antenna gain is consistent with its size and frequency
Estimate the gain, beamwidth or effective aperture of a dish of known diameter
Find the minimum far-field range for antenna measurements or Friis validity
Size an antenna for a required gain at a given frequency
Returns a PhysicalViolationError dict if any input violates physics or the claimed gain exceeds the aperture limit.
Inputs: frequency_hz (required), diameter_m, aperture_area_m2, claimed_gain_dbi, +1 more. Returns: object
noise_floorA
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_1G_2) + ..., the receiver's effective input noise temperature T_e = T_0(F-1) with T_0 = 290 K (the IEEE reference at which noise figure is defined, independent of temperature_k), and receiver sensitivity S_min = k_B*(T_A + T_e)BSNR_required, which reduces to N_floor + NF + SNR_required when temperature_k = 290 K.
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.
Inputs: bandwidth_hz (required), stages, temperature_k, required_snr_db. Returns: object
radar_ambiguityA
Calculate pulse-Doppler radar ambiguity limits and validate range/velocity claims.
Computes the maximum unambiguous range R_ua = c / (2 * PRF), the first blind speed lambda * PRF / 2, the unambiguous velocity v_ua = +/- lambda * PRF / 4 (Doppler within +/- PRF/2), the Doppler shift f_d = 2 * v_r / lambda of a target (closing velocity positive) and whether it aliases, and — when a pulse width is given — the duty cycle, eclipsing minimum range c * tau / 2 and the range resolution c * tau / 2 (or c / (2 B) when a compressed bandwidth is supplied). Reports the range-Doppler dilemma invariant R_ua * v_ua = c * lambda / 8, which no choice of PRF can beat (Skolnik, Introduction to Radar Systems, Ch. 2-3; Richards, Fundamentals of Radar Signal Processing, Ch. 1, 3, 5).
Use this tool when you need to:
Find the unambiguous range and velocity coverage of a given PRF
Check whether a claimed unambiguous range or velocity is possible at that PRF
Check whether a claimed range resolution is possible for the pulse/bandwidth
Determine whether a target's Doppler will alias or fall at a blind speed
Expose claims that silently violate the range-Doppler dilemma
Returns both human-readable summary and machine-readable JSON with all intermediate values. Returns a PhysicalViolationError dict if any input violates physics (PRF <= 0, tau <= 0, duty cycle >= 1) or a claim exceeds its limit.
Inputs: prf_hz (required), frequency_hz (required), bandwidth_hz, pulse_width_s, +4 more. Returns: object
radar_rangeA
Calculate maximum monostatic radar detection range and validate range claims.
Computes the radar range equation R_max = [P_t * G^2 * lambda^2 * sigma / ((4*pi)^3 * S_min * L)]^(1/4) for a monostatic radar (same antenna for transmit and receive). Validates that claimed detection ranges do not exceed the theoretical maximum. Catches the common fourth-root fallacy where LLMs incorrectly state that doubling transmit power doubles radar range (it only increases range by a factor of 2^(1/4) = 1.19x).
Use this tool when you need to:
Calculate the maximum detection range of a radar system
Validate whether a claimed radar detection range is physically achievable
Determine minimum detectable signal power for a radar receiver
Check if radar performance claims account for the R^4 path loss
Verify that RCS assumptions are reasonable for the target class
Returns both human-readable summary and machine-readable JSON with all intermediate values. Returns a PhysicalViolationError dict if any input violates physics or the claimed range exceeds R_max.
Inputs: rcs_m2 (required), frequency_hz (required), peak_power_w (required), antenna_gain_dbi (required), +6 more. Returns: object
rf_link_budgetA
Calculate a complete RF link budget using the Friis transmission equation.
Computes free-space path loss (FSPL), received power, and validates antenna gains against the physical gain limit. The hard bound is G_max = max((piD/lambda)^2, (ka)^2 + 2ka) with k = 2pi/lambda and a = D/2, i.e. the larger of the eta = 1 aperture value and Harrington's bound for an antenna enclosed in a sphere of diameter D (numerically the Harrington value, which converges to the aperture value for D >> lambda and matters for electrically small antennas, D < lambda); a claimed gain above it is rejected. A claimed gain above the typical-efficiency value eta*(pi*D/lambda)^2 (eta = 0.55 by default) but below the physical bound is accepted with a warning. Negative losses are rejected (conservation of energy).
Use this tool when you need to:
Estimate received signal strength for a wireless link
Validate whether a claimed link budget is physically achievable
Check if antenna gain claims are consistent with antenna dimensions
Compute free-space path loss at a given frequency and distance
Returns both human-readable summary and machine-readable JSON with all intermediate values. Returns a PhysicalViolationError dict if any input violates physics.
Inputs: distance_m (required), frequency_hz (required), tx_power_dbm (required), rx_antenna_gain_dbi (required), +6 more. Returns: object
shannon_hartleyA
Calculate Shannon-Hartley channel capacity and validate throughput claims.
Computes the theoretical maximum data rate C = B * log2(1 + SNR) for an AWGN channel. If a claimed throughput is provided, validates it against this limit. Any claim exceeding the Shannon limit is a physical impossibility.
Use this tool when you need to:
Calculate maximum achievable throughput for a given bandwidth and SNR
Validate whether a throughput claim is physically possible
Determine spectral efficiency limits
Check if a modulation/coding scheme claim is realistic
Returns a PhysicalViolationError dict when a claim exceeds the Shannon limit.
Inputs: bandwidth_hz (required), snr_db, snr_linear, claimed_throughput_bps. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Each tool targets a distinct RF physics equation family, and the descriptions are detailed enough to disambiguate most cases. However, rf_link_budget and antenna_gain both validate antenna gain claims against the same physical aperture/Harrington limits, so an agent could initially be uncertain which tool to call for a gain-checking task.
All six tool names follow a consistent snake_case noun phrase pattern: rf_link_budget, shannon_hartley, noise_floor, radar_range, antenna_gain, radar_ambiguity. There is no mixing of conventions or vague verbs, so the naming is predictable and readable.
Six tools is well-scoped for a specialized RF/physics validation server. Each tool covers a distinct and substantial calculation area, and none feel redundant or unnecessary.
The tool set covers the major RF physical-layer validation surfaces: link budgeting, channel capacity, noise floor and sensitivity, radar range, antenna gain limits, and pulse-Doppler ambiguity. Within the apparent domain of validating RF claims against physics, there are no obvious dead ends or missing core operations.
Maintenance
Related MCP Connectors
AI-callable calculators and engineering models with real formulas. No hallucinated math.
Jailbreak-proof AI guardrails. Automated Reasoning SMT solver, not an LLM. ZK proofs included.
Deterministic runtime safety for AI agents: scan PII, gate tool actions, verify LLM output.
33 tools that make AI write, implement, and verify intent against explicit, testable constraints.
Related MCP Servers
- AlicenseCqualityDmaintenanceBlueMouse is the "Prefrontal Cortex" for LLMs. It uses a 180k+ failure pattern database to validate code logic before execution, acting as a rigorous Quality Gate to prevent hallucinations and unsafe operations.8109AGPL 3.0
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to verify physical admissibility of S-parameter models and coupling matrices, preventing reasoning about impossible physics.Apache 2.0
- AlicenseAqualityBmaintenanceProvides a local offline physics/chemistry simulation bench covering 12 domains, where plain-language intents are converted into simulation blueprints, solved with deterministic proof metrics, and exported to STL/OBJ — no cloud, API key, or LLM required.480 npm1Apache 2.0
- AlicenseAqualityBmaintenanceHeadron is a model-independent AI agent verification and governance platform. It analyzes code changes, determines required checks, enforces policies, and generates proof-carrying receipts. Integrates with MCP, CI/CD, GitHub Actions, and coding agents. Unrun checks never count as passes. Its local, read-only MCP server requires no account or API key.14233 PyPI4MIT