Security audit

NFS-e Parser — Brazilian invoice field extraction

Security checks for vulnerabilities and agentic risk

Overview

The skill is purpose-aligned for local invoice parsing, but it asks users to fetch and run mutable remote parser code without integrity checks while handling sensitive invoice data.

Review or pin the external parser before installing, preferably to a specific commit with a verified checksum, and run OCR/parsing in a limited workspace without unrelated credentials or files. Treat NFS-e PDFs and extracted text as sensitive financial data; use cloud OCR only with explicit approval and appropriate privacy/compliance review.

A.I.G

Vulnerability Patterns

Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Insecure DependenciesIntroduces malicious components through unsafe dependency sources
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
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (2)

T03 · Remote Payload Retrieval and Execution

Warning

Location: SKILL.md:40
Finding: Mutable Remote Parser Is Retrieved and Executed Without Integrity Verification
Content: View full analysis
Remediation: View remediation

T08 · Insecure Dependencies

Warning

Location: SKILL.md:37
Finding: Python Dependencies Are Installed Without Reproducible Version and Integrity Pins
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The skill references Google Document AI as an OCR upstream for invoice processing without warning that invoice PDFs and extracted text may contain sensitive financial and personal data and would be transmitted to a third-party cloud service. In the context of NFS-e documents, this can lead to unintended disclosure of taxpayer identifiers, addresses, billing details, and retention data, especially if an agent selects the service automatically based on accuracy or cost tradeoffs.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.