Security audit

OpenDataLoader PDF

Security checks for vulnerabilities and agentic risk

Overview

This PDF parsing skill is useful and coherent, but users should review it because sensitive documents may be sent to remote or AI-backed services without clear privacy warnings.

Install only in a virtual environment or container, avoid unpinned upgrades when possible, and treat all converted PDF content as sensitive. Use localhost or a trusted authenticated backend for hybrid mode, do not point OPENDATALOADER_HYBRID_URL at an untrusted service, enable sanitization for untrusted PDFs, and be aware that the RAG example using OpenAI embeddings may send extracted text to a third-party provider.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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

Findings (2)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:12
Finding: Unpinned Third-Party Packages Installed from a Mutable Package Source
Content: View full analysis
Remediation: View remediation

T09 · Insecure Skill Coding Practices

Warning

Location: references/api-reference.md:5
Finding: Prompt-Injection Sanitization Is Disabled by Default for Untrusted PDF Content
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 (2)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The skill advertises OCR, hybrid processing, and AI-enrichment features for PDFs but does not clearly disclose whether document contents, images, or extracted text are transmitted to a local service, remote backend, or third-party AI component. Users may process sensitive PDFs such as medical reports or invoices under the assumption of local-only handling, creating a real confidentiality and compliance risk if data is sent elsewhere.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The documentation instructs users to run a hybrid backend and also supports a remote backend via the OPENDATALOADER_HYBRID_URL environment variable, but it does not warn that PDF contents may be transmitted to another service for processing. This can lead users to send sensitive document contents to a local or remote server without understanding the privacy, confidentiality, and data-handling implications.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.