Origin Pro MCP

• Outcome: Review
• Latest audit: May 27, 2026, 1:40 PM
• Version:
  0.2.1

Overview

This Origin Pro automation skill is mostly transparent about its purpose, but it needs Review because it can run arbitrary Origin scripts and discard unsaved work without built-in safeguards.

Install only if you intentionally want an AI agent to control an active Origin Pro session. Keep important Origin projects saved, avoid broad implicit use, review any LabTalk before execution, and prefer a version that disables implicit invocation or adds confirmation gates for new_project and raw LabTalk.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 94% confidence
• Finding: The skill instructs users to run shell commands, install packages, register an MCP server, and execute Python-based tests, but it does not declare corresponding permissions or capabilities. In a skill ecosystem, this mismatch weakens transparency and consent boundaries: users or orchestrators may not realize the skill can drive shell, environment-variable, and file-writing behavior, increasing the risk of unintended local changes or command execution.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The manifest enables implicit invocation for a skill that controls a local MCP server backed by Windows COM automation, but it does not define clear trigger constraints or narrow usage boundaries. This increases the chance the agent will invoke a powerful local automation capability in unintended contexts, potentially causing unauthorized Origin Pro actions, data manipulation, or execution of risky automation flows without explicit user confirmation.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: This function passes arbitrary strings directly into Origin's LabTalk interpreter with no validation, restriction, or user-confirmation boundary in this module. In the context of an MCP skill that automates Origin Pro, upstream user input could reach this sink and trigger unintended Origin actions, file access, command execution features exposed by LabTalk/Origin, or destructive modification of projects and data.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The tool builds LabTalk commands by interpolating user-controlled inputs such as workbook name, sheet name, column numbers, and function name directly into command strings passed to execute_labtalk. If LabTalk metacharacters or unexpected values are accepted, an attacker could inject additional LabTalk commands and gain unauthorized control over Origin operations, which is especially risky because this skill is explicitly designed to drive a powerful automation interface.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: This tool directly exposes execution of arbitrary LabTalk provided as a raw string and returns the full script in the response, with no validation, allowlisting, confirmation step, or scope restriction. In the context of an MCP tool that an agent may call on a user's behalf, this creates a powerful command-execution surface inside Origin that could modify projects, overwrite data, trigger file operations, or be abused through prompt injection or unsafe tool use.

Missing User Warnings

High

• Confidence: 96% confidence
• Finding: The new_project tool directly calls Origin's NewProject(), which closes the current project without saving and provides no confirmation, warning, or safeguard. In an MCP/agent context, this creates a real integrity and availability risk because a mistaken, ambiguous, or maliciously induced tool call can silently destroy unsaved work in the user's active Origin session.

VirusTotal

VirusTotal findings are pending for this skill version.