Matlab Bridge
PassAudited by ClawScan on May 10, 2026.
Overview
This skill does what it claims—runs MATLAB for analysis and plotting—but users should treat it like a local code-execution tool and keep an eye on saved outputs.
Install only if you intend to let OpenClaw run local MATLAB code. Review MATLAB scripts or generated code for sensitive tasks, keep outputs in a known directory, and clean saved .mat/workspace files if they contain private data.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
MATLAB code run through this skill can perform calculations, create files, and potentially access or modify local files within the user's permissions.
The skill builds and runs a MATLAB script from caller-supplied code in a local MATLAB process. This is the advertised bridge behavior, but it means the skill can execute powerful local MATLAB commands.
def execute(self, code: str, ...): ... full_code = self._build_wrapper_code(code, temp_name, save_workspace) ... '-batch', f"cd('{temp_dir}'); {temp_name}"Use it only for trusted MATLAB code or scripts, review generated code before execution when possible, and avoid running prompts that ask MATLAB to modify unrelated files.
Data used during MATLAB analysis may remain on disk after the task, including sensitive experimental or business data if it was loaded into the workspace.
The skill saves the MATLAB workspace by default, which can persist variables and processed data from a task into local .mat files.
def execute(self, code: str, save_workspace: bool = True, ...): ... mat_file = fullfile(output_dir, '{temp_name}_workspace.mat'); save(mat_file);Disable workspace saving for sensitive tasks when possible and periodically review or delete files in the MATLAB output directory.
The skill may fail or behave unexpectedly if MATLAB is not installed or not on PATH, and the platform may not warn about that automatically.
The registry metadata does not machine-declare MATLAB as a required binary even though SKILL.md instructs users to have MATLAB installed and available on PATH. This can reduce automatic compatibility checks, but the prerequisite is disclosed in the documentation.
Required binaries (all must exist): none ... Install specifications: No install spec — this is an instruction-only skill.
Confirm MATLAB is installed, licensed, and on PATH before using the skill; consider adding MATLAB as a declared required binary in metadata.