Audit History
google-analytics-automation - 5 audits
Version comparison
Capability and finding changes across audited versions, newest first.
| Version | Date | Result | Review items | Change vs previous |
|---|---|---|---|---|
| v5 Latest | Jul 24, 2026, 12:34 AM | 1 confirmed | 1 | No capability change |
| v4 | Jul 8, 2026, 11:47 AM | 1 confirmed | 1 | No capability change |
| v3 | Jul 7, 2026, 01:17 AM | No confirmed findings | 1 | External commands |
| v2 | Jun 30, 2026, 04:32 PM | 1 confirmed | 0 | Network access |
| v1 | Feb 25, 2026, 03:27 AM | No confirmed findings | 0 | Baseline |
Jul 24, 2026, 12:34 AM
All 60 command-execution alerts are false positives caused by Markdown backticks and code fences. The skill makes an expected third-party MCP connection and delegates Google OAuth through Rube.
Confirmed security concerns (1)
Capability review items (1)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (50)
🌐 Network access (1)
Jul 8, 2026, 11:47 AM
The Ruby and shell execution alerts are false positives caused by Markdown backticks around tool names, parameters, and workflow examples. The remaining confirmed concern is a low-risk external dependency: users must configure Rube MCP and complete Google OAuth for Google Analytics access.
Confirmed security concerns (1)
Capability review items (1)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (60)
🌐 Network access (1)
Jul 7, 2026, 01:17 AM
Most static alerts are false positives from Markdown inline code that names MCP tools, GA4 parameters, or code fences. The only confirmed issue is the documented external Rube MCP endpoint, which introduces a low third-party network and OAuth handling risk. No prompt injection or hidden malicious intent was found in SKILL.md.
Capability review items (1)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (60)
🌐 Network access (1)
Jun 30, 2026, 04:32 PM
Static analysis flagged many Ruby shell backticks, weak crypto, and reconnaissance patterns, but review shows Markdown code spans, front matter, or GA4 metadata guidance. The confirmed risk is use of an external Rube MCP endpoint and Google OAuth to access analytics data. No prompt injection, command execution, credential exfiltration, or malicious intent was found.
Confirmed security concerns (1)
Static false positives ignored (1)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
Risk Factors
🌐 Network access (1)
Feb 25, 2026, 03:27 AM
All static findings are false positives. The 92 'external_commands' detections are MCP tool names (e.g., GOOGLE_ANALYTICS_LIST_ACCOUNTS), not shell executions. The hardcoded URL is a legitimate MCP server endpoint. No cryptographic algorithms or system reconnaissance present. This is a legitimate Google Analytics automation skill using Rube MCP.