Security audit

excel-format-optimizer

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Excel formatting helper that edits workbook presentation and saves an optimized copy, with some dependency and formatting-default cautions but no evidence of hidden or malicious behavior.

Install only if you are comfortable with the agent installing openpyxl from the named package mirror and making broad visual formatting changes to Excel files. Keep the original workbook, confirm the target font and formatting plan, and approve any optional LibreOffice formula recalculation separately.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: The skill instructs the agent to install a dependency from an external mirror over the network, including a user-specific local environment path. That introduces unnecessary supply-chain and environment-modification risk for a formatting skill, because the behavior goes beyond processing the provided workbook and can fetch untrusted code from outside the task boundary.

Context-Inappropriate Capability

Medium

Confidence: 82% confidence
Finding: The documentation expands the skill into invoking another skill/script path for formula recalculation with LibreOffice, which is outside the stated Excel-formatting scope. Cross-skill execution increases attack surface and can enable unexpected document processing behavior not obvious from the skill's advertised purpose.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The trigger language is broad enough to match common requests like making a table 'look better,' which can cause this skill to activate in situations the user did not specifically intend. Over-broad invocation is dangerous because the skill writes modified Excel files and may make sweeping formatting changes to important documents without sufficiently precise consent.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill begins by describing workbook modification behavior but does not provide a prominent upfront warning that it will alter formatting across the file and save a new output workbook. For document-editing skills, lack of clear pre-action disclosure can lead to unintended mass changes to user data presentation and workflow disruption.

Natural-Language Policy Violations

Medium

Confidence: 76% confidence
Finding: Mandating a specific locale-dependent font without user opt-in can alter document appearance, layout, and compatibility in ways the user may not expect, especially for multilingual or branded workbooks. In context this is less severe than code-execution issues, but it is still a risky default because it can cause rendering regressions and unintended semantic formatting changes at scale.

Static analysis

No suspicious patterns detected.