ClawHub
Back to skill

Security audit

Glitchward Shield

Security checks across malware telemetry and agentic risk

Overview

This skill appears to do what it says: send prompts to Glitchward's external shield API to check for prompt-injection patterns.

Before installing, confirm you are comfortable sending prompts or prompt excerpts to Glitchward for scanning and store GLITCHWARD_SHIELD_TOKEN as an environment secret. Avoid routing sensitive or regulated data through the service unless your policy permits that processing.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
This markdown file states that user input is validated through Glitchward's Shield API and shows a POST request containing prompt text, but it does not explicitly warn users that their prompts will be sent to a third-party service. Because this behavior can affect user data privacy, the skill description should disclose it plainly.

Instruction Override

High
Category
Prompt Injection
Content
{
      "category": "instruction_override",
      "severity": "critical",
      "pattern": "ignore all previous instructions"
    }
  ]
}
Confidence
80% confidence
Finding
ignore all previous instructions

YARA rule 'agent_skill_prompt_injection_hidden_instructions': Prompt injection or hidden instructions embedded in AI agent skill text [agent_skills]

High
Category
YARA Match
Content
# Glitchward LLM Shield — OpenClaw Skill

Prompt injection detection for AI agents. Scan prompts through a 6-layer detection pipeline before they reach your LLM.
Confidence
80% confidence
Finding
LLM; AI agent; LLM; llm; AI agent; LLM; system prompt; LLM; LLM; llm; ignore all previous instructions; ignore all previous instructions

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.prompt_injection_instructions

Prompt-injection style instruction pattern detected.

Warn
Code
suspicious.prompt_injection_instructions
Location
README.md:46