ClawHub
Back to skill

Security audit

Pilot Task Chain

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Pilot orchestration helper that chains agent tasks, with expected data-sharing risk but no hidden payload or destructive behavior in the artifact.

Install this only if you trust the Pilot tooling and the agents you will chain. Treat intermediate results as potentially sensitive or prompt-injectable, and review, summarize, or redact them before sending them to agents that can store, publish, delete, or modify important data.

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 (1)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The workflow examples explicitly pass one agent's output directly into another agent's task prompt without any validation, redaction, or warning about cross-agent data sharing. In an orchestration skill, this can propagate sensitive data, secrets, or untrusted content across trust boundaries and increase the chance of data disclosure or prompt-injection-style instruction smuggling between agents.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.