ClawHub
Back to skill

Security audit

RSS采集器免费版

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed RSS-to-local-SQLite helper with expected network fetching and local storage, though its trigger wording should be read narrowly.

Install only if you want an agent to fetch RSS feeds over the network and maintain a local SQLite article database. Keep its use limited to the RSS archive it creates, review commands before disabling or deleting sources, and back up data/rss_fetcher.db if the archive matters to you.

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
84% confidence
Finding
The trigger condition is broad enough to activate the skill for generic database or storage-management requests unrelated to RSS collection. In an agent environment with exec and local file/database access, this can cause overreach into unintended tasks, leading to unauthorized database reads/writes or accidental data modification when the wrong skill is selected.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill performs network fetching and persistent local storage, but the summary/triggering guidance does not clearly foreground those side effects. In agent settings, insufficient disclosure can lead users to invoke the skill without understanding that it will contact external URLs and write to disk, which raises privacy, compliance, and integrity risks.

Missing User Warnings

Low
Confidence
83% confidence
Finding
The instructions include direct database queries and operational changes such as disabling sources, but they do not clearly warn about local state changes or the possibility of modifying persistent data. In an agentic environment, this omission can cause users to authorize actions without realizing they may alter configuration or database contents, increasing the risk of unintended data loss or system-state changes.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.