SciMath MCP
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@SciMath MCP Calculate the derivative of x^3*sin(x)"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
SciMath MCP
SciMath is a local, read-only Model Context Protocol server for reliable engineering and scientific computation. It moves arithmetic and mathematical manipulation out of an LLM's reasoning and into established numerical, symbolic, unit, and uncertainty libraries.
It is designed specifically for tool-calling agents—including smaller quantized local models—so the model chooses and explains a calculation while mature libraries execute it reproducibly.
What it provides
Exact and arbitrary-precision arithmetic with SymPy.
Unit-aware formulas and dimensional validation with Pint.
Equations, calculus, and exact linear algebra.
Descriptive statistics, regression, confidence intervals, and Welch t-tests.
First-order propagation of independent measurement uncertainties.
Physical constants from SciPy's bundled CODATA table.
Structured MCP output containing exact values, approximations, normalized inputs, metadata, and warnings.
Every expression is parsed by an AST allowlist. Python eval, attribute access, imports,
comprehensions, indexing, assignment, and arbitrary function calls are not available. Each
calculation runs in a disposable subprocess with input, result-size, complexity, and time limits.
Related MCP server: Verified Technical Solver
Tools
Tool | Purpose |
| Arithmetic, scientific functions, high precision, and unit-aware formulas |
| Compatible unit conversion and dimensional validation |
| Real or complex algebraic equations and systems |
| Derivatives, integrals, limits, and series |
| Exact matrix operations plus condition-number warnings |
| Summaries, correlation, regression, intervals, and Welch t-tests |
| Measurement uncertainty with optional units |
| CODATA constant lookup with uncertainty and unit conversion |
Run locally
Prerequisites: Python 3.12 or newer and uv.
uv sync --all-groups
PYTHONPATH=src uv run python -m scimath_mcpInstall a stable executable:
uv tool install --force /absolute/path/to/scimath-mcpThen register the same stdio executable in each client:
codex mcp add scimath -- "$HOME/.local/bin/scimath-mcp"
claude mcp add --scope user scimath -- "$HOME/.local/bin/scimath-mcp"
opencode mcp add scimath -- "$HOME/.local/bin/scimath-mcp"Verify with codex mcp get scimath, claude mcp get scimath, and opencode mcp list. The
executable requires no network access. After changing the source, rerun the uv tool install
command to refresh the installed wheel.
See Client setup for complete JSON/TOML examples and troubleshooting.
Expression syntax
Use explicit multiplication (2*x). Both ** and ^ mean exponentiation. Common functions
include sqrt, cbrt, exp, ln, log, log10, sin, cos, tan, inverse and hyperbolic
trigonometry, factorial, gamma, erf, floor, ceil, min, and max. Constants include
pi, tau, e, phi, and complex I in unitless calculations.
Put dimensionful values in the variables object:
{
"expression": "M*c/I",
"variables": {
"M": "12.4 kN*m",
"c": "75 mm",
"I": "8.7e-6 m^4"
},
"output_unit": "MPa"
}Reliability boundary
SciMath makes the execution of a supplied formula reproducible. It cannot prove that an LLM selected the correct physical model, sign convention, branch, domain, units, or statistical assumptions. Consumers should show normalized inputs and warnings for consequential work.
Documentation
Client setup — Codex CLI, Claude Code, and OpenCode.
Tool reference — inputs, operations, limits, and examples.
Architecture — trust boundaries and request lifecycle.
Design notes — reviewed community projects and design decisions.
Security — threat model, reporting, and numerical safety.
Contributing — development workflow and release checks.
Tests
uv run pytest
uv run ruff check .
uv build --clearLicense
MIT
Available Tools
8 toolscalculateARead-onlyIdempotent
Evaluate one arithmetic/scientific expression safely. Use explicit * and variables.
For dimensional work, pass original unit-bearing variable strings (for example 12.4 kN*m)
and always set output_unit; never add manual conversion factors to the expression. Returns
exact and high-precision numeric forms when unitless.
Inputs: expression (required), precision, variables, angle_unit, +1 more. Returns: Stable, machine-readable result shared by every computation tool.
calculusARead-onlyIdempotent
Differentiate, integrate, take a limit, or expand a series symbolically. For definite
integrals provide both bounds; for limits and series provide point. order selects derivative
order or the number of series terms.
Inputs: operation (required), expression (required), order, point, +5 more. Returns: Stable, machine-readable result shared by every computation tool.
convert_unitsARead-onlyIdempotent
Convert a dimensionful value such as 32 degF, 120 km/hour, or 5 psi to a compatible
target unit. Dimensional mismatches are errors rather than guessed conversions.
Inputs: value (required), to_unit (required), precision. Returns: Stable, machine-readable result shared by every computation tool.
linear_algebraARead-onlyIdempotent
Perform exact matrix and linear-algebra operations on matrices up to 25x25. Supply
other_matrix for multiplication or as the right-hand side of solve. String entries such as
1/3 preserve exact values. Condition-number results include numerical-stability warnings.
Inputs: matrix (required), operation (required), precision, other_matrix. Returns: Stable, machine-readable result shared by every computation tool.
physical_constantARead-onlyIdempotent
Look up a physical constant from SciPy's bundled CODATA table by common alias or name. Returns its value, unit, standard uncertainty, source metadata, and an optional compatible unit conversion. Examples: c, G, hbar, Boltzmann constant, electron mass.
Inputs: query (required), precision, output_unit. Returns: Stable, machine-readable result shared by every computation tool.
propagate_uncertaintyBRead-onlyIdempotent
Propagate independent standard uncertainties through an expression. Each measurement has value, uncertainty, and optional unit. Returns nominal value, standard uncertainty, relative uncertainty, and a coverage interval; use output_unit when the result is dimensionful.
Inputs: expression (required), measurements (required), precision, angle_unit, +2 more. Returns: Stable, machine-readable result shared by every computation tool.
solveARead-onlyIdempotent
Solve up to eight algebraic equations for explicitly named variables. Each equation may
use left = right or a zero-valued expression. Returns every solution in exact and numerical
form; choose the real or complex domain explicitly.
Inputs: equations (required), variables (required), domain, precision. Returns: Stable, machine-readable result shared by every computation tool.
statisticsARead-onlyIdempotent
Compute descriptive statistics, Pearson correlation, linear regression, a Student-t confidence interval for a mean, or an independent Welch t-test. Provide data_y for paired correlation/regression or the second t-test sample.
Inputs: data (required), operation (required), data_y, precision, +1 more. Returns: Stable, machine-readable result shared by every computation tool.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 8 tools
Each tool has a clear, non-overlapping purpose: calculate evaluates expressions, solve handles equations, calculus does symbolic differentiation/integration, linear_algebra handles matrix operations, etc. Even superficially similar tools like calculate and calculus are distinct in scope.
All names use lowercase with underscores for multi-word tools, but the pattern is mixed: some are verbs (calculate, solve), some are noun phrases (linear_algebra, physical_constant). This is mostly consistent but deviates from a uniform verb_noun style.
With 8 tools, the set is well-scoped for scientific mathematical computation. Each tool covers a major area (arithmetic, units, equations, calculus, linear algebra, statistics, uncertainty, constants) without redundancy or bloat.
The tool surface covers the core scientific computing workflow: expression evaluation, unit conversion, equation solving, symbolic calculus, matrix operations, statistical analysis, uncertainty propagation, and constants lookup. Minor gaps like numeric root-finding or optimization exist, but the set is comprehensive for typical math/science use.
Maintenance
Related MCP Connectors
Exact arithmetic, algebra, calculus, physics, chemistry and finance with a checkable formula.
Exact fraction math for AI agents. No floating-point drift. CODATA 2022 constants and units.
Free deterministic tools for math, data transforms, dates, text, statistics, finance, and units.
AI-callable calculators and engineering models with real formulas. No hallucinated math.
Related MCP Servers
- AlicenseBqualityBmaintenanceProvides secure mathematical computation capabilities including expression evaluation, symbolic math (derivatives, simplification), matrix operations, statistics, and unit conversion, with multi-tier acceleration through WebAssembly and WebWorkers for high-performance calculations.718 npmISC
- AlicenseNot gradedqualityDmaintenanceProvides deterministic tools for symbolic math, equation solving, relation verification, reasoning structure audits, and validated CAD primitives, returning evidence artifacts for every result.MIT
- AlicenseAqualityBmaintenanceProvides a safe scientific runtime for agents with typed math operations including calculus, algebra, statistics, unit conversion, and more via MCP tools.4Apache 2.0
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to perform safe sandboxed expression evaluation, symbolic calculus such as derivatives, simplification, equation solving and integration, statistical analysis, and JSON-native matrix operations. Everything is computed statelessly through the Model Context Protocol over stdio, HTTP, or SSE.MIT