ClawHub
Back to skill

Security audit

Praesidia

Security checks across malware telemetry and agentic risk

Overview

Praesidia is a coherent API integration, but it can send user content to Praesidia and make account-backed guardrail changes without consistently requiring explicit user confirmation.

Install only if you intend to use Praesidia with a least-privilege API key. Verify PRAESIDIA_API_URL, avoid using it for generic security discussions unless you want Praesidia API calls, and require explicit approval before sending sensitive text for validation or changing any guardrail configuration.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The manifest description contains very broad trigger phrases such as generic security, compliance, discovery, and 'is this agent safe?' prompts, which can cause the skill to be invoked in situations where the user did not clearly intend to call an external verification service. In this skill's context, unintended invocation is more dangerous because many flows lead to authenticated API requests and may expose user queries, agent identifiers, or organization context to a remote service.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill instructs sending arbitrary user-provided content to a remote guardrail validation endpoint but does not require a clear user-facing disclosure or consent step before transmission. This is dangerous because users may paste sensitive prompts, personal data, regulated data, or proprietary information expecting local analysis, while the skill silently forwards that content to an external service.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.