mcp-numpy
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| np_arrayB | Create a NumPy array from a list. |
| np_zerosB | Create an array of zeros. |
| np_onesA | Create an array of ones. |
| np_fullC | Create an array filled with a constant value. |
| np_arangeC | Create an array with evenly spaced values within a given interval. |
| np_linspaceB | Create an array with evenly spaced numbers over a specified interval. |
| np_eyeC | Return a 2D identity array. |
| np_diagC | Create a diagonal array or extract the diagonal of an array. |
| np_reshapeB | Give a new shape to an array without changing its data. |
| np_transposeA | Reverse or permute the axes of an array. |
| np_concatenateB | Join a sequence of arrays along an existing axis. |
| np_splitC | Split an array into multiple sub-arrays. |
| np_tileA | Construct an array by repeating the input array the given number of times. |
| np_repeatC | Repeat elements of an array. |
| np_squeezeB | Remove single-dimensional entries from the shape of an array. |
| np_flattenB | Return a flattened copy of the array. |
| np_sumB | Sum of array elements over given axis(es). |
| np_meanA | Compute the arithmetic mean along the specified axis. |
| np_stdB | Compute the standard deviation along the specified axis. |
| np_varC | Compute the variance along the specified axis. |
| np_minA | Return the minimum of an array or minimum along an axis. |
| np_maxA | Return the maximum of an array or maximum along an axis. |
| np_argminB | Return the indices of the minimum values along an axis. |
| np_argmaxA | Return the indices of the maximum values along an axis. |
| np_dotB | Compute the dot product of two arrays. |
| np_matmulB | Matrix product of two arrays. |
| np_crossB | Compute the cross product of two arrays. |
| np_traceB | Return the sum along the main diagonal of the array. |
| np_cumsumB | Return the cumulative sum of the array along a given axis. |
| np_cumprodA | Return the cumulative product of the array along a given axis. |
| np_diffB | Calculate the n-th discrete difference along the given axis. |
| np_invA | Compute the (multiplicative) inverse of a matrix. |
| np_detC | Compute the determinant of an array. |
| np_eigA | Compute the eigenvalues and eigenvectors of a square array. |
| np_svdC | Singular Value Decomposition. |
| np_solveB | Solve a linear matrix equation, or system of linear equations. |
| np_linalg_normB | Matrix or vector norm. |
| np_randC | Random values in a given shape. |
| np_randnB | Return a sample (or samples) from the "standard normal" distribution. |
| np_randintB | Return random integers from low (inclusive) to high (exclusive). |
| np_random_choiceC | Generates a random sample from a given array. |
| np_shuffleA | Modify a sequence in-place by shuffling its contents. |
| np_percentileC | Compute the q-th percentile of the array elements. |
| np_quantileC | Compute the q-th quantile of the array elements. |
| np_histogramB | Compute the histogram of a set of data. |
| np_correlateB | Cross-correlation of two 1-dimensional sequences. |
| np_corrcoefB | Return Pearson product-moment correlation coefficients. |
| np_addB | Element-wise addition of two arrays. |
| np_subtractA | Element-wise subtraction of two arrays. |
| np_multiplyA | Element-wise multiplication of two arrays. |
| np_divideC | Element-wise division of two arrays. |
| np_powerB | Element-wise exponentiation of array elements. |
| np_modA | Element-wise modulo of two arrays. |
| np_sqrtB | Return the non-negative square root of an array element-wise. |
| np_absA | Calculate the absolute value of array elements. |
| np_expA | Calculate the exponential of all elements in the array. |
| np_logA | Natural logarithm, element-wise. |
| np_log10B | Base-10 logarithm, element-wise. |
| np_sinA | Trigonometric sine, element-wise. |
| np_cosB | Trigonometric cosine, element-wise. |
| np_tanA | Trigonometric tangent, element-wise. |
| np_arcsinB | Inverse sine, element-wise. |
| np_arccosB | Inverse cosine, element-wise. |
| np_arctanB | Inverse tangent, element-wise. |
| np_sinhB | Hyperbolic sine, element-wise. |
| np_coshA | Hyperbolic cosine, element-wise. |
| np_tanhC | Hyperbolic tangent, element-wise. |
| np_shapeB | Return the shape of an array. |
| np_ndimA | Return the number of dimensions of an array. |
| np_sizeA | Return the total number of elements in an array. |
| np_dtypeB | Return the dtype of an array. |
| npastypeC | Copy of the array, cast to a specified type. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 72 tools
Each tool targets a distinct NumPy function with clear, specific descriptions. No two tools perform the same operation; even closely related functions like np_percentile and np_quantile are differentiated by definition and use case.
Most tools follow a consistent 'np_operation' pattern. The only exception is 'npastype' which lacks an underscore, and 'np_linalg_norm' uses a submodule prefix while other linear algebra tools (np_eig, np_inv, etc.) do not. Overall, the pattern is clear and predictable.
72 tools is excessive for an MCP server, making selection difficult for an agent. While the server aims to cover a broad library like NumPy, the count far exceeds the typical well-scoped range (3-15) and even the 'too many' threshold (25+).
The tools cover core array operations, math, statistics, linear algebra, and random sampling. However, common functions like sorting (np_sort), unique (np_unique), where (np_where), and fft are missing, representing notable gaps for a numpy-oriented server.