ClawHub
Back to skill

Security audit

Geo analytics & AI visibility

Security checks across malware telemetry and agentic risk

Overview

This is a coherent read-only analytics skill, but it handles non-public site analytics and a long-lived local access token.

Install only if you are comfortable giving the skill read access to one site's non-public analytics reports. Keep the saved access token private, prefer restricted local permissions or a credential store, and revoke it in the Vtag console when it is no longer needed. Avoid pasting raw realtime, session, or probe-event data into places where visitor behavior or business queries should not be exposed.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (4)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation advises storing a long-lived bearer access token in a local file but does not warn that the token is effectively a credential tied to a site and should be protected like a secret. If saved insecurely, other local users, malware, backups, logs, or tooling could recover the token and gain ongoing read access to non-public analytics data.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
`/probe-events` is documented as returning raw sampled records including `answer_text` and `sources`, which may contain sensitive business queries, proprietary prompts, URLs, or other user-derived content. Although the file notes not to compute metrics from this endpoint, it does not warn that this is sensitive raw content requiring restricted access, minimization, or careful handling, increasing the chance of overexposure through the skill.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The realtime endpoint exposes near-live event data with visitor-level fields such as `visitor_id`, event timestamps, page paths, navigation source, device, and location. Even if the data is first-party analytics, documenting this endpoint without an explicit privacy warning or constraints encourages use of highly linkable telemetry that could enable tracking, re-identification, or misuse in dashboards and downstream tools.

Missing User Warnings

Low
Confidence
82% confidence
Finding
`/ivt/session-events` provides single-session drilldown data, and the related IVT report includes session identifiers and potentially truncated user-agent data. Session-level event sequences can reveal browsing behavior and support correlation with other logs, so the lack of a clear warning about sensitive session-level exposure increases the risk of privacy misuse or overcollection.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.