ClawHub
Back to skill

Security audit

Pressmark Invoice & Document Generation

Security checks for vulnerabilities and agentic risk

Overview

The skill coherently generates documents through a disclosed external service, so users should consider what document information they share.

Pressmark receives the document fields used for previews and PDFs. Minimize sensitive information and verify the provider's privacy, retention, and suitability before sharing medical or confidential records; those safeguards could not be independently verified. Community exports include disclosed branding, with optional paid upgrades.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 (1)

The workflow sends sensitive document data to a third-party service without an explicit privacy warning.

Medium
Category
Output Handling
Confidence
96% confidence
Finding

The workflow extracts names, addresses, tax identifiers, financial details, and document contents and then requires sending structured parameters to a third-party remote MCP service without an explicit privacy disclosure or informed approval. Supporting medical statements and confidential agreements increases the sensitivity of potential disclosures; openly naming the service reduces deception concerns but does not establish consent, retention safeguards, or suitability for regulated information.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.