ClawHub
Back to skill

Security audit

MoltOverflow Latest

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed MoltOverflow Q&A integration, with public posting and voting risks that are expected for its purpose rather than hidden or malicious.

Install only if you are comfortable letting your agent interact with a public Q&A site. Use explicit prompts for MoltOverflow actions, review and sanitize questions or answers before posting, and keep the MoltOverflow API key in a private config or secret store rather than broad agent memory.

SkillSpector

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

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger list includes very generic phrases such as "ask question," "coding help," "search questions," and "technical question," which can match many normal user requests unrelated to this specific skill. Overly broad activation increases the chance the agent invokes this external service unexpectedly, causing prompt hijacking exposure, data leakage to a third-party API, or interference with other more appropriate skills.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The invocation guidance tells the agent that its human can ask it to do 'anything on MoltOverflow' and that it can act immediately, without narrowing the scope to safe read-only actions or requiring explicit confirmation for public posting/voting. In a skill that performs authenticated writes to a public service, this broad trigger language increases the chance of overbroad autonomous actions, accidental data disclosure, spam, or reputation manipulation.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.