ClawHub
Back to skill

Security audit

Pre-Publish Fact Checker

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent fact-checking helper that uses a disclosed Prismfy API key and optional reminder hook, with privacy and credential-handling cautions but no artifact-backed malicious behavior.

Install if you are comfortable sending the claims you verify to Prismfy using your API key. Prefer a revocable or scoped key, avoid putting secrets in shared dotfiles, and enable the hook only if you want future agent sessions to receive fact-checking reminders.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill instructs the agent/user to run shell scripts and shell commands, but the manifest does not declare corresponding permissions or capabilities. This weakens transparency and policy enforcement because a reviewer or runtime may underestimate what the skill can execute, increasing the chance of unintended command execution or unsafe integration.

Tp4

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding
The declared purpose is limited to claim verification, but the skill also performs external API access, quota/account inspection, and installs/enables a hook that affects agent behavior outside a single verification run. This mismatch is security-relevant because hidden or under-disclosed behaviors can surprise operators, expand data exposure to third parties, and create persistence-like effects in the local agent environment.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The script sends user-supplied claim text to an external third-party API and can also retrieve account data, but it provides no user-facing disclosure, consent gate, or sensitivity check before transmission. In a claim-verification skill, drafts may contain unpublished, confidential, or regulated information, so silent exfiltration to an external service creates a real data-handling risk in context.

YARA rule 'backdoor_persistence': Backdoor persistence with malicious payloads (shell commands, SSH key injection, hidden root users) [malware]

High
Category
YARA Match
Content
To keep it after restart:
```bash
echo 'export PRISMFY_API_KEY="ss_live_your_key_here"' >> ~/.bashrc
source ~/.bashrc
```
Confidence
86% confidence
Finding
echo 'export PRISMFY_API_KEY="ss_live_your_key_here"' >> ~/.bashrc

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.