PowerPoint PPTX Layout Export

• Outcome: Pass
• Latest audit: Jun 15, 2026 at 1:38 AM UTC
• Version:
  0.1.0

Overview

This is a PowerPoint helper skill with no executable payload; its main risk is overly broad invocation wording, not hidden or malicious behavior.

This skill is reasonable to install for PowerPoint/PPTX automation. Use it intentionally on decks you choose, keep backups of important presentations, and watch for accidental activation on generic presentation discussions because some trigger examples are broader than ideal.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (8)

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The trigger sentences are extremely broad, malformed, and generic enough that unrelated PowerPoint or workflow requests could accidentally invoke the skill. In an agent environment, unintended invocation can cause incorrect tool routing, unexpected file handling, or unnecessary processing of presentation artifacts, which increases the chance of downstream security or reliability issues.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger phrases are extremely broad and partially templated, making the skill likely to activate on loosely related PowerPoint requests rather than explicit user intent. In an agent system, over-broad activation can cause unintended tool selection, misrouting, or execution of a skill on untrusted content, which increases the chance of unsafe or incorrect actions.

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The skill description embeds a very broad demand statement and says to use the skill whenever the user mentions loosely related topics like PowerPoint, PPTX, slide layout, or chart export. This can cause the agent to invoke the skill outside its intended boundary, leading to incorrect routing, over-collection of context, or unsafe downstream automation on files the user did not intend to modify.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The keyword list is highly generic for common office-work requests and lacks exclusion conditions or contextual qualifiers. Because these terms appear in many benign conversations about presentations, the skill may activate too often and steer the agent into unnecessary file-processing or code-generation workflows.

Vague Triggers

Low

• Confidence: 84% confidence
• Finding: The sample trigger phrases use open-ended wording like 'I need' and include repeated broad demand text rather than precise task boundaries. This weakens trigger discrimination and can train or encourage an invocation layer to match on vague user phrasing that is unrelated to the specialized PPTX automation use case.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The default prompt and broad description make the skill eligible for implicit invocation on a wide range of generic PowerPoint requests, without clearly constraining when it should be used. In an agentic system, this can cause over-triggering of a powerful file-manipulation skill, increasing the chance of unintended document modification, export, or processing outside the user's intended scope.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The trigger sentence begins with a very broad, everyday phrase ('Help me ...'), which can cause the skill to activate outside its intended PowerPoint/PPTX-specific scope. In an agent environment, overly generic triggers increase the chance of accidental routing, prompt interception, or unintended execution when unrelated user requests happen to partially match the phrase.

Vague Triggers

Medium

• Confidence: 96% confidence
• Finding: This trigger sentence is ambiguous and underspecified, so the agent may invoke the skill for requests that only loosely resemble the intended use case. Because this skill can influence document generation and transformation workflows, ambiguous activation broadens attack surface and can lead to misfires, unsafe tool selection, or processing of unintended content.

VirusTotal

64/64 vendors flagged this skill as clean.