100Critical
The skill allows arbitrary command execution via Bash
Behavior Analysis
Description verified — behavior matches claim
Claims to do
CMS Platform Detection: You are a CMS detection specialist. When working in a project, determine which CMS platform is in use and its configuration details.
Actually does
This skill uses `Read`, `Grep`, and `Glob` to scan local project files such as `.csproj`, `sitecore.json`, `appsettings.json`, `module.config`, and specific directories. It identifies the CMS (Sitecore, Umbraco, or Optimizely) and its configuration by looking for specific files, NuGet package references, and configuration strings within these files. It does not contact any external URLs or execute arbitrary `Bash` commands beyond file system operations.
Skill Info
Nametwofoldtech-dakota/plugin-cms-toolkit/cms-detect
Registrygithub
Version27fbabdd4251
PURLpkg:github/twofoldtech-dakota/plugin-cms-toolkit@27fbabdd4251?skill=cms-detect
Stars1
Source
SHA-2561b00e87f4861...
SHA-1d9cd14d5d4a7...
MD507079e257ccd...
TLSHc061a7376049...
Size3,220 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
Assessments (1)
AI Agent Traps ↗Arbitrary Command Execution via Bashcriticalcommand execution
The skill explicitly allows the 'Bash' tool, enabling arbitrary command execution. This poses a critical risk for data exfiltration, system modification, and other malicious activities if the agent is compromised or instructed to misuse this capability.
100% confidenceline 5
allowed-tools: Read, Grep, Glob, Bash
Broad File System Access for Reconnaissancehighreconnaissance
The combination of 'Read', 'Grep', 'Glob', and 'Bash' grants broad access to the file system. While intended for CMS detection, these tools can be misused to enumerate sensitive files, configurations, and system information beyond the project scope.
100% confidenceline 5
allowed-tools: Read, Grep, Glob, Bash
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/twofoldtech-dakota/plugin-cms-toolkit/cms-detect)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/twofoldtech-dakota/plugin-cms-toolkit/cms-detect"><img src="https://mondoo.com/ai-agent-security/api/badge/github/twofoldtech-dakota/plugin-cms-toolkit/cms-detect.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/twofoldtech-dakota/plugin-cms-toolkit/cms-detect.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow