Pubblue

Security checks across malware telemetry and agentic risk

Overview

This skill appears purpose-aligned, but it asks users to run an externally sourced CLI daemon that can publish content, keep live network sessions, store an API key, and access an agent workspace.

Review before installing. Verify the pubblue npm package and source provenance, avoid running npx or @latest in sensitive environments, use a sandboxed PUBBLUE_CONFIG_DIR, keep the API key out of shared logs and workspaces, and only use --public or live sessions with content and peers you intend to expose.

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: 89% confidence
Finding: The skill includes concrete API key setup commands and reveals the default on-disk config location, but does not warn users that the key is a sensitive secret requiring careful handling. In an agent-tooling context, this increases the chance of credential leakage through shell history, pasted commands, logs, shared workspaces, or inadvertent file exposure.

Missing User Warnings

Medium

Confidence: 83% confidence
Finding: The skill promotes publishing and live browser-linked visualization, including a documented `--public` mode, without clearly warning that content may be exposed to external parties. Because this skill is specifically designed to share agent output, users may unintentionally publish sensitive data or enable live viewing of information that should remain private.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal