Security audit

Mineru Extract

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to perform its stated MinerU document-processing job, but users should understand that documents or URLs may be sent to a third-party service and output files may be created locally.

Install only if you are comfortable sending the documents or URLs you choose to MinerU or related remote APIs. Avoid confidential, regulated, or internal-only files unless you have approval, and run it from a working directory where newly created output files will not overwrite anything important.

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

Confidence: 95% confidence
Finding: The skill explicitly states that output is always written to the user's current working directory, but it does not warn about file creation, naming, or overwrite behavior. In an agent context, this can unexpectedly modify local files or collide with existing files, creating integrity and data-loss risk even if the intent is not malicious.

Missing User Warnings

Medium

Confidence: 98% confidence
Finding: The skill is designed to send user-provided local files and URLs to external MinerU endpoints, including unauthenticated and authenticated remote APIs, but it provides no clear privacy or data-transmission warning. This is dangerous because users may unknowingly upload sensitive documents or internal URLs to a third-party service, causing confidentiality and compliance exposure.

Static analysis

No suspicious patterns detected.