Pdf Smart Tool Cn 1.1.0

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:43 PM UTC
• Version:
  1.0.0

Overview

This is an instruction-only PDF utility with sensitive document features, but no hidden code, persistence, or data exfiltration was found.

Install only if you are comfortable processing PDFs through this agent. Use it on documents you are authorized to modify, confirm file lists before batch actions, avoid unnecessary password or certificate entry, and keep backups before decryption, signing, watermark removal, or format conversion.

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
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The trigger words listed in the description include very common terms such as "PDF", "转换", and "识别", which are likely to match ordinary user conversation outside the intended workflow. Overly broad triggers can cause unintended activation of the skill in contexts involving sensitive documents, increasing the chance of accidental processing or disclosure.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill promotes OCR, conversion, and document analysis features but does not warn users that document contents may be exposed to the tool during processing. Because PDFs often contain contracts, IDs, invoices, and other sensitive data, lack of disclosure can lead to users submitting confidential material without informed consent.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The decryption and password-handling sections instruct users to enter PDF passwords and remove protections, but they do not include any warning about credential sensitivity, authorization requirements, or safe handling of decrypted output. This creates elevated risk because users may expose secrets, process protected files they are not authorized to decrypt, or leave decrypted copies insufficiently protected.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The skill advertises trigger words such as “PDF”, “转换”, and “识别”, which are generic terms likely to appear in normal conversation. This can cause accidental activation of a powerful document-processing skill in unrelated contexts, increasing the chance of unintended file operations or prompting users into sensitive workflows without clear intent.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill description highlights sensitive capabilities including OCR, digital signing, encryption/decryption, watermark removal, and batch processing, but does not warn about privacy, legal, or destructive consequences. In this context, users may upload confidential PDFs, remove protections, overwrite document integrity, or process many files at once without informed consent or safety checkpoints.

VirusTotal

63/63 vendors flagged this skill as clean.