ClawHub
Back to skill

Security audit

Reposit - Collective Intelligence for AI Agents

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Reposit integration that may contact an external community backend during coding, with data-scrubbing instructions and sharing confirmation enabled by default.

Install only if you are comfortable with an agent sending sanitized problem descriptions to Reposit during coding. Keep REPOSIT_AUTO_SHARE disabled unless you explicitly want automatic publication, review shared content carefully, and avoid using the skill on highly sensitive private work unless you trust the Reposit backend and MCP package.

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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The README describes very broad automatic triggers such as searching when 'encountering errors' or 'starting complex work', which can cause the skill to activate in many normal workflows without clear user awareness. In a skill that sends data to an external community service, overbroad activation increases the chance of unnecessary disclosure of prompts, errors, code fragments, or sensitive work context.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The README says search triggers automatically and that search works without authentication, but it does not warn that error messages or work context may be transmitted to an external service. This creates a real risk of unintentional data exfiltration, especially because agents may include stack traces, file paths, proprietary code, secrets, or customer data in search queries during normal operation.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill instructs the agent to invoke external search automatically under broad conditions such as any unfamiliar error, any non-trivial problem, or when seeking a better way. In this context, automatic outbound queries create a real risk of unintended data disclosure and unreviewed third-party interaction, especially because the skill is designed to transmit problem descriptions to a remote backend.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.