ClawHub

Echo Repeater

Security checks across malware telemetry and agentic risk

Overview

This is a low-capability echo skill with a visible but confusing extra output sentence, not evidence of malware or high-impact access.

Install only if you are comfortable with a toy echo skill that may append an extra holiday sentence. Do not use it for exact-output tests, message integrity checks, or repeating secrets, passwords, API keys, personal data, or content where added text would cause problems.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
98% confidence
Finding
The skill’s core purpose is to echo the user's input exactly, but the documented output format injects additional text: "Happy New Year!". This creates instruction/data integrity drift and can cause the skill to emit content the user did not provide, which is especially risky in testing, message-routing validation, or workflows that rely on exact round-trip behavior.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger criteria are broad enough to match trivial or common phrases, increasing the chance this skill activates outside narrowly intended testing scenarios. An over-broad echo skill can unintentionally repeat sensitive or unsafe content in contexts where another safer or more task-appropriate skill should have handled the request.

Natural-Language Policy Violations

Medium
Confidence
97% confidence
Finding
The output instructions require adding an English phrase the user did not request, violating user intent and the stated exact-echo semantics. Forced extra output can pollute downstream tests, alter quoted content, and create integrity issues in systems expecting faithful reproduction of input.

Ssd 3

Medium
Confidence
90% confidence
Finding
A verbatim echo skill will reproduce whatever the user supplies, including secrets, tokens, personal data, or harmful text. In a testing-oriented skill, this risk is amplified because users may intentionally paste sensitive strings to verify routing, causing unnecessary redisclosure in logs, transcripts, or shared channels.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal