Security audit

SlideSpeak

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a straightforward SlideSpeak API integration, with privacy cautions around sending selected files and prompts to an external service.

Install only if you are comfortable sending selected presentation prompts, local documents, and slide configuration data to SlideSpeak. Keep SLIDESPEAK_API_KEY secret, avoid uploading confidential or regulated documents unless approved, and use REMOVE and webhook subscription commands deliberately.

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: 95% confidence
Finding: The skill explicitly instructs users to upload local documents to a third-party API but does not warn that the file contents will leave the local environment and be transmitted to SlideSpeak. This can lead users to unintentionally send sensitive, proprietary, or regulated data to an external service, especially because the workflow presents upload as a normal step without privacy caveats.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The API reference explicitly supports uploading user documents to a third-party service and also supports webhook callbacks, but it provides no warning that potentially sensitive file contents, metadata, or generated presentation data leave the local environment. In an agent skill context, this omission can cause users or integrators to unknowingly transmit confidential business documents or personal data to an external processor and external callback endpoint.

Missing User Warnings

Medium

Confidence: 80% confidence
Finding: The slide editing API includes a destructive REMOVE operation but does not warn that it deletes content from an existing presentation or clarify whether the action is reversible. In an agent-driven workflow, ambiguous destructive operations increase the risk of accidental data loss if a model or user issues the wrong edit request.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The upload command reads an arbitrary local file path and transmits the file contents to a third-party API, but the script provides no explicit user-facing warning, confirmation, or path restrictions. In an agent setting, this increases the risk of unintended exfiltration of sensitive local documents if a user or upstream workflow invokes upload on confidential files.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The generation commands send user-provided text, document identifiers, or JSON config content to an external service without making the data transfer and privacy implications explicit. In a security-sensitive agent environment, users may provide confidential business content assuming it stays local, creating an avoidable data disclosure risk.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.