Back to skill

Security audit

pdf-tool

Security checks for vulnerabilities and agentic risk

Overview

This is a simple PDF utility description with expected local file handling and no hidden execution, persistence, or data-sharing behavior.

Before installing, confirm you are comfortable with a local PDF helper that may read document contents and write output files. Avoid using sensitive PDFs unless you trust the runtime environment, and choose non-conflicting output filenames.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

Most operational instructions, parameter descriptions, and tables are written in Chinese, while the skill does not state that language selection is optional or region-specific. This can constitute a language/locale policy issue because the skill effectively defaults to a specific language without explicit user opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The feature list advertises extracting text from PDFs, including contracts in the examples and use cases, but does not warn that extracted text may contain sensitive or confidential information. Because the markdown describes behavior that can expose user data in plain text, a privacy-oriented warning should be present.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

This markdown file describes operations like merge, split, and watermark that create or modify output PDF files, but it does not warn users that running the skill will write files to disk and may overwrite existing paths if reused. For a file-manipulating skill, a brief user-facing warning about output file effects is expected under the markdown-specific warning criterion.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.