lhx111

• Outcome: Review
• Latest audit: May 29, 2026 · 9:25 PM UTC
• Version: 1.0.0

Overview

This appears to be a real spreadsheet helper, but it silently adds a persistent LibreOffice macro to the user profile and runs LibreOffice on local workbooks.

Install only if you are comfortable with a local spreadsheet tool that can modify and save workbooks, run LibreOffice headlessly, and add a persistent LibreOffice macro to your user profile. Keep backups of important spreadsheets and avoid using it on untrusted or sensitive workbooks unless you can sandbox LibreOffice or inspect the generated macro.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (7)

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

if timeout_cmd:
            cmd = [timeout_cmd, str(timeout)] + cmd
    
    result = subprocess.run(cmd, capture_output=True, text=True)
    
    if result.returncode != 0 and result.returncode != 124:  # 124 is timeout exit code
        error_msg = result.stderr or 'Unknown error during recalculation'

• Confidence: 95% confidence
• Finding: result = subprocess.run(cmd, capture_output=True, text=True)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 94% confidence
• Finding: The skill clearly instructs reading and writing spreadsheet files and executing a shell command (`python recalc.py ...`), yet no declared permissions are present. This creates a capability/permission mismatch that can lead to silent file system access or command execution without explicit policy review or user awareness.

Vague Triggers

Medium

• Confidence: 83% confidence
• Finding: The description is extremely broad and could cause the skill to activate for many generic spreadsheet-related requests, including ones where users may not expect file modification or formula recalculation workflows. Over-broad routing increases the chance the skill is invoked in contexts involving sensitive local files or unintended write operations.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The examples demonstrate creating and modifying files directly but do not require confirmation, disclosure, or caution about overwriting data. In an agent setting, that can normalize unsafe write behavior and lead to unintended changes to user files or templates.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill instructs subprocess-style execution of `python recalc.py` without requiring any warning, path validation, or provenance check for the script. This is dangerous because it normalizes local code execution, and if `recalc.py` is replaced or supplied from an untrusted location, the agent could run attacker-controlled code.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: Stating that LibreOffice setup is automatic implies system-level installation or configuration side effects without user notification. Silent environment modification is risky in secure or managed systems and may violate user expectations or platform policy.

Missing User Warnings

Medium

• Confidence: 98% confidence
• Finding: The script writes a persistent LibreOffice application-level macro into the user's profile without prompting or clearly warning the user. Modifying office-suite startup/application state is dangerous because it creates lasting trust-boundary changes, may overwrite existing macros, and can alter future behavior outside this script's immediate execution.

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.