Langchain Skill

Security checks across malware telemetry and agentic risk

Overview

The skill appears useful, but it handles uploaded files, document search, and web/tool access without enough privacy and data-handling boundaries.

Install only if you are comfortable giving the skill access to the documents and queries you provide. Avoid uploading confidential, regulated, or credential-containing files unless the publisher documents where data is processed, whether embeddings are stored, how to delete indexed content, and when external services are used.

SkillSpector

By NVIDIA

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: 93% confidence
Finding: The skill advertises web search, calculator, and file-reading capabilities but does not disclose what data may be sent to external services or read from local/user-provided sources. In an agent context, this can lead users to unknowingly expose sensitive prompts, file contents, or browsing-derived data to third-party providers or tools.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The documentation promotes PDF upload and RAG over uploaded documents without warning users how document contents are processed, stored, embedded, or potentially transmitted to model or vector services. This creates a real risk of sensitive document leakage, especially if users assume uploads remain local or ephemeral.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal