Security audit

Web Crawling API

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward Just Serp web-crawling skill, with the main privacy caveat that requested URLs are sent to that external provider.

Install only if you are comfortable sending target webpage URLs and fetched content through Just Serp API. Avoid confidential, internal-only, signed, tokenized, or personal-data-bearing URLs unless your organization has approved that provider and use case.

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 (5)

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill sends a user-supplied target URL and the user's API key to the external service api.justserpapi.com without any in-code disclosure, confirmation, or guardrails. In an agent setting, this can expose sensitive internal URLs, tokens embedded in URLs, or user browsing targets to a third party, creating a privacy and data-handling risk even though the code does not appear overtly malicious.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill sends a user-provided URL to a third-party crawling service, which means user inputs and any embedded sensitive query parameters are transmitted outside the host environment. Because the operation descriptions do not warn about this external disclosure or constrain what URLs may be submitted, users or downstream agents may unknowingly expose private internal links, tokens in URLs, or sensitive browsing targets.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The operation accepts a user-supplied URL and sends it to an external crawling provider, but the documentation does not warn users that their input will be transmitted off-platform. This can cause unintended disclosure of sensitive URLs, internal endpoints, or query parameters if a user assumes the fetch happens locally or within a trusted boundary.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: This endpoint converts a user-provided URL to Markdown via an external service, but the spec omits any warning about third-party transmission of the URL. Without that disclosure, users may provide sensitive or private links that are exposed to the external crawler and potentially logged or retained by that provider.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The rendered HTML endpoint forwards user-specified URLs to an external rendering/crawling service, yet the generated docs do not disclose that data flow. In skill contexts, this is more dangerous because users may treat the tool as a generic fetch utility and unknowingly submit private resources, creating a confidentiality and trust risk.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.