29 tools from the Mcp Numpy MCP Server, categorised by risk level.
View the Mcp Numpy policy →np_arange Create array with range 2/5 np_array Create a NumPy array 2/5 np_concatenate Concatenate arrays 2/5 np_det Matrix determinant 2/5 np_diag Create diagonal array 2/5 np_eig Eigenvalues and eigenvectors 2/5 np_eye Create identity matrix 2/5 np_flatten Flatten array 2/5 np_full Create array filled with value 2/5 np_histogram Histogram 2/5 np_inv Matrix inverse 2/5 np_linalg_norm Matrix/vector norm 2/5 np_linspace Create evenly spaced array 2/5 np_ones Create ones array 2/5 np_rand Random floats 2/5 np_randint Random integers 2/5 np_randn Random normal 2/5 np_random_choice Random choice 2/5 np_repeat Repeat elements 2/5 np_reshape Reshape array 2/5 np_shuffle Shuffle array 2/5 np_solve Solve linear system 2/5 np_split Split array 2/5 np_svd Singular value decomposition 2/5 np_tile Tile array 2/5 np_transpose Transpose array 2/5 np_zeros Create zeros array 2/5 npastype Type conversion 2/5 The Mcp Numpy MCP server exposes 29 tools across 2 categories: Write, Destructive.
Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the Mcp Numpy server.
Mcp Numpy tools are categorised as Write (28), Destructive (1). Each category has a recommended default policy.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept